1 Rmd Description

This is Part 4 for the project focused on the extension and multiverse analysis of a publication in Development and Psychopathology 2020

The sections are split across Aim 1 and Aim 2. In Aim 1 we conduct the replication analyses from the original study, which focused on the mediating role of puberty (parent report) on the direct association between the Family Environment factor (independent variable) and the brain outcomes (5 brain outcomes: Amygdala volume, ACC thickness, ACC area, and Left & Right Amyg-Cing Network resting state connectivity. Note, we do not use the sixth brain outcome, ACC FA, due to the significant change in preprocessing of DWI data between release 1 and 2). In Aim 2 we extend these results by conducting the multiverse analyses for model permutations across IV variables and the mediator, with all brain outcomes unchanged.

Note: the option for this rmarkdown is code_folding: hide, so the code is hidden. Please select Code –> Show All Code to unhide all code blocks. Otherwise, in each section you can simply click the Code button to show the respective code chunk.

2 Packages

2.1 Required packings, install

if (!require("pacman")) install.packages("pacman")
pacman::p_load(janitor, cowplot, dplyr, readr, devtools, table1, MatchIt, kableExtra, corrplot, multiplex, MplusAutomation,lavaanPlot, semPlot, lavaan, parameters, specr, MASS, Hmisc, patchwork)

2.2 Load packages

At each instance, load packages.

# Load packages
library(tidyverse) # piping, descriptives, etc
library(kableExtra) # for table formatting
library(lavaan) # running lavaan Factor models +  mediation 
library(lavaanPlot)
library(parameters) # reporting lavaan parameters in table
library(specr)
library(cowplot)
library(MASS)
library(glue)
library(corrplot)
library(Hmisc)
library(patchwork)

2.3 Importing Final data

Here we import the data that was created during Part 2 & 3, which includes our variables and factor scores being used in the models below.

data <- read.csv("../../stage1/ABCD_wFactorScores_2022_07_11.csv")

minor correction, the preregistration missed the step which averaged the parent/child reported PDS score & the z-scored parental income and education variable. So we create these variables below.

# averaging the parent and child reported PDS scores
data <- data %>% 
  mutate(avg_puberty = rowMeans(dplyr::select(.,c(p_puberty,y_puberty)), 
                             na.rm = F)) # Use F instead of T, this way NA is provided in cases of missing values on yth or par report.

# z-scoring & averaging parental income and education
data$Par_Inc_z <- scale(data$Par_Inc, center = T, scale = T)
data$Par_Edu_z <- scale(data$Par_Edu, center = T, scale = T)

data <- data %>% 
  mutate(Avg_IncomeEduc = rowMeans(dplyr::select(.,c(Par_Inc_z,Par_Edu_z)), 
                             na.rm = F))

Let’s also rename the variables that we use so they are easily interpretable and easier to use the numerical values the models below.

data <- data %>% 
  rename(
      # IV: renaming our factor scores
      FamEnv_Fact = FamEnv_Score_r, Demo_Fact = Demo_Score_r,
      Par_Fact = Par_Score_r, Child_Fact = Child_Score,
      # IV: renaming variables
      FES_youth = FES_y_mean_r, CRPBI = CRPBI_mean, PMON = PMON_y_mean,
      FES_parent = FES_p_mean_r,
      # outcome brain
      Amygdala_vol = Amyg_Vol_z, ACC_CA = ACC_Area_z, 
      ACC_CT = ACC_Thick_z, 
      L_AmygCing_rest = Lamyg_CingOperc, R_AmygCing_rest = Ramyg_CingOperc,
      # covariates
      
      )

data = data %>% 
  mutate(race_r = case_when(race == 'White' ~ 1,
                          race == 'Black' ~ 2,
                          race == 'Hispanic' ~ 3,
                          race == 'Asian' ~ 4,
                          race == 'Other' ~ 5),
         sex_r = case_when(sex == 'F' ~ 0,
                           sex == 'M' ~ 1)
         )

Consistent with the preregistered table 1 the variables and their associated variable names that we will focus on for the replication^ and multiverse analyses are as follows

Indepdendent variables (IV)

  • Family Environment Factor: FamEnv_Fact^
  • Demographic Factor: Demo_Fact
  • Child Factor: Child_Fact
  • Parent Factor: Par_Fact
  • FES Youth: FES_youth (reverse scored)
  • FES Parent: FES_parent (reverse scored)
  • Parental Monitoring: PMON
  • Parental Acceptance: CRPBI
  • z-scored Parental Income/Education: Avg_IncomeEduc

Mediating Variables (M)

  • Par Reported PDS: p_puberty^
  • Yth reported PDS: y_puberty
  • Averaged Par/Yth PDS: avg_puberty

Outcome Variables (DV)

  • Bilat. Amygdala SV: Amygdala_vol^
  • Bilat. ACC CT: ACC_CT^
  • Bilat. ACC CA: ACC_CA^
  • Left Amygdala-CON rsfMRI: L_AmygCing_rest^
  • Right Amygdala-CON rsfMRI: R_AmygCing_rest^

Covariates in all models

  • Age: Age
  • Race/Ethnicity: race_r
  • Sex: sex_r

For purposes of coherence in abbreviations as we report in the manuscript, we removed underscores (e.g., FamEnv_Fact –> FamEnv Fact) and renamed some variables as follows. FES_youth = FES Yth; FES_parent = FES Par; CRPBI = Par Accept; Avg IncomeEduc = Avg IncEdu; p_puberty = PDS Par; y_puberty = PDS Par; avg_puberty = PDS ParYth; L/R AmygCing_rest = L/R AmygCON.

3 Multiverse Function

Here we define our multiverse function, where model inputs include:

  • List of IVs (X)
  • List of DVs (Y)
  • List of Mediators (M)
  • List of covariates (Cov)
  • Specified lavaan model
  • Aim 1: this can be used for our Five* model replication, as we run five models
    • Fam | PDS Par | Amygdala Vol
    • Fam | PDS Par | ACC CT
    • Fam | PDS Par | ACC CA
    • Fam | PDS Par | L Amyg ~ CON FC
    • Fam | Par Puberty | R Amyg ~ CON FC

*Note: Fam | Par Puberty | ACC FA we do not include the FA measure given the changes in preprocessing for DWI data between the releases which impacted the underlying estimates of FA.

  • Aim 2: this is particularly useful for our 100+ multiverse estimate

3.1 Function code

Some helpful notes regarding this function. There are three steps. The function requires input of: X = list of independent variables, Y = list of dependent variables, M = list of mediators, Cov = list of covariates (note, for this project we do not have a list, but in stead use constants of Age, Race, Sex), df A datamrame that contains our variables (columes) and participants (rows), lavaan_model = this is the syntax specifying the mediation model, model_boot = the number of bootstraps to run in the mediation model that calculates significance for our effects.

  1. Step 1 simply takes in the X, Y, M list + Cov1, Cov2, Cov2 constants and creates the permutation of variable combinations. This is simply the product of the list of items in X, Y, M, Covs. It calculates this product and spits out the combinations used.

  2. In step 2 we use apply() to each row to run all variable permutations (matrix rows) for our lavaan::sem() model. The apply function takes each row of variables (stored in var_perumtations), sends them into the function variable combos, subsets our df by these variables and creates these column names w/ that data. On this data, lavaan function is used to fit the specified model. The mediation model uses the permutations of X, Y, M + constant covariates.

  3. After Step 2 is complete, we will get a list of total permutations saved to the data_rep. We can access the model, variable and fit statistics via different ways. Here, we use lappy() to loop through this list, and send each instance to function(x) where x is each iteration of output in data_rep. We use the parameters package to save the standardized model parameters to a data.frame, and we save the variable names for each model Using the location of each variable and parameter name and estimate value, we extract this estimate and save it to a data.frame. The variables that are relevant here are:

    • Model’s variables permutations across: X, Y, M
    • Number of observations for the model
    • Number of parameters fit for the model
    • The effect, e.g., A, B, Direct, Indirect & total effect paths
    • Beta estimate, associatied 95% CI, conf.high, conf.low, and the p.value

We combine across rows to get these values in a signal dataframe, then relabel the column names so we have the output structure that is comparable to the specr() output style to use w/ plot_spercs() function(s).

in the pre-registered code there was an error in which estimates were extract. In the pre-registration model_parameters(fit, standardize = FALSE) when it should have been model_parameters(fit, standardize = TRUE)

mediation_specr <- function(X, Y, M, Cov1, Cov2, Cov3, df, lavaan_model, model_boot) {

## Step 1
  # Creating variable permutations for lists in X, Y, M, Cov
var_permutations <- as.matrix(expand.grid(X = X, 
                                          Y = Y, 
                                          M = M, 
                                          Age = Cov1,
                                          Race = Cov2,
                                          Sex = Cov3))

# Check the number of permutations that are created basic ont he input strings
total_permutes <- length(X)*length(Y)*length(M)*length(Cov1)*length(Cov2)*length(Cov3)
print(paste0("Total Permutations for Current Mediation Models: ", total_permutes))


# Step 2
# Create data.frame of output using model pre
##  the var_permutations are run across the matrics into the function.


data_rep <- 
  apply(X = var_permutations, MARGIN = 1, function (combos) {
    require(lavaan)
    permuted.df <- df[combos] # pull data that only has variables we're interested in
    
    # selection variable names to replace in model_fit()
    X = as.character(combos[1]) 
    Y = as.character(combos[2])
    M = as.character(combos[3])
    Age = as.character(combos[4])
    Race = as.character(combos[5])
    Sex = as.character(combos[6])
    
    # we use "glue()" here in the model sytnax to example {variables} with the values assigned above. 
    mediation_model <- glue('
    # Direct Effect (X->Y), c - path
    {Y} ~ c*{X} + {Age} + {Race} + {Sex}
    
    # Meidation (X -> M), a path
    {M} ~ a*{X} + {Age} + {Race} + {Sex}
    
    # Mediation (M -> Y), b path 
    {Y} ~ b*{M} 
    
    # Indirect Effect (a*b)
    ind := a*b
    
    # Total Effect
    total := c + (a*b)
    ')
    
    # Herre we run and save the model parameters to "fit"
    fit <-lavaan::sem(model = mediation_model, 
                      data = permuted.df,
                      se = "bootstrap",
                      bootstrap = model_boot, 
                      mimic = "Mplus"
                      )

# Step 2
# Extra the variables we will be using in subsequent steps
require(parameters)

model_out = data.frame(model_parameters(fit, standardize = TRUE)) 
med_vars = lavNames(fit)
spec_data <- data.frame("X"  = as.character(med_vars[3]), 
                       "Y" = as.character(med_vars[1]), 
                       "M" = as.character(med_vars[2]), 
              "Direct_estimate" = model_out[1,4], "Direct_std.error" = model_out[1,5],
              "Direct_conf.low" = model_out[1,6],  "Direct_conf.high" = model_out[1,7], 
              "Direct_p.value" = model_out[1,9],
           
              "Apath_estimate" = model_out[5,4], "Apath_std.error" = model_out[5,5],
              "Apath_conf.low" = model_out[5,6],  "Apath_conf.high" = model_out[5,7], 
              "Apath_p.value" = model_out[5,9],
              "Bpath_estimate" = model_out[9,4], "Bpath_std.error" = model_out[9,5],
              "Bpath_conf.low" = model_out[9,6],  "Bpath_conf.high" = model_out[9,7],
              "Bpath_p.value" = model_out[9,9],
           
              "Indirect_estimate" = model_out[16,4], "Indirect_std.error" = model_out[16,5],
              "Indirect_conf.low" = model_out[16,6],  "Indirect_conf.high" = model_out[16,7], 
              "Indirect_p.value" = model_out[16,9],
              
              "Total_estimate" = model_out[17,4], "Total_std.error" = model_out[17,5],
              "Total_conf.low" = model_out[17,6],  "Total_conf.high" = model_out[17,7], 
              "Total_p.value" = model_out[17,9]) 
  # add model observations & parameters N
  spec_data$Observations = fitMeasures(fit)[21]; spec_data$N_Parameters = fitMeasures(fit)[1]
  
  return(spec_data)
}) %>% 
  bind_rows() %>%  # combined by rows the lists from lapply
  gather(key = "Effect", value = "Coefficient", Direct_estimate:Total_p.value) %>% # we take all colum names Direct:Total_p and save the variable names, i.e., Direct, to the key "Effect" and the value to "Coefficient"
  separate(col = "Effect", into = c("Effect","Type"), sep = "_", extra = "merge", fill = "right") %>% # because we want to combine this data to similar format as we safe our thijseen file, we want to separate all variables with '_', that way the first value 'Direct' or 'Apath' is in "Effect" and the type of value, i.e., SE or lower CI, is in 'Type' column. We fill these to the right
  spread(key = "Type", value = "Coefficient") # now that we have the type column, we spread this data out from long to wide
  

}

4 Aim 1

4.1 Thijssen et al effects

We extract the Betas and Standard errors from the original paper (i..e, Figure 1, Table 4 - 6), and calculate the 95% CI for the Direct Effect, A path, B path using Beta +/- 1.96 * SE. We save these in a .csv file to use in comparisons here. Given that there was not sufficient information to calculate these metrics based on information provided in the original paper, Dr. Thijssen provided the SE/95% CI values for total & indirect effects.

Of note: In the original pre-registration, we had a typo in the effect preregistered for the R/L Amyg CON indirect effect which didn’t reflect of those in Thijssen et al. 2021 reanalysis post philips scanner reported error by consortium, whereby we pre-registered an effect that was not the one the authors re-analyzed in their data after excluding later reported issue with philips scanner’s preprocessing. This did not change any subsequent inferences here.

Thij_eff <- read.csv("./Thijssenetal_Effects.csv") %>% 
  rename("estimate" = Beta, "std.error" = SE, 
         "conf.low" = lower_95CI, "conf.high" = upper_95CI, 
         "p.value" = pval) 


# Remove the ACC_FA, as we are not using this in replication
Thij_eff <- Thij_eff %>% 
  filter(Brain != "ACC FA")

Here we plot the Beta estimates and any associated 95% Confidence Interval for each path type: a path, b path, indirect (a*b), direct (c`) and total (c) across the five brain DVs.

color_1 <- cbPalette <- c("#000000", "#E69F00", "#56B4E9", "#009E73", "#F0E442", "#0072B2")# six colours for six factors

Thij_eff %>% 
  ggplot(aes(x = Brain, y = estimate, colour = Brain)) +
  #geom_errorbar(aes(ymin=len-ci, ymax=len+ci), colour="black", width=.1, position=pd) +
  geom_errorbar(aes(ymin = conf.low, ymax = conf.high, colour = Brain), 
                 width = .1, alpha = .5) +
  geom_point(fill = "white", shape = 21) +
  labs(y = "Beta Estimates") +
  scale_colour_manual(values = color_1)+
  facet_wrap(~Effect, scale = "free")+
  theme_minimal()+
  theme(axis.title.x = element_blank(), axis.ticks.x = element_blank(), 
        axis.line.x = element_blank(), axis.text.x = element_blank())

Thij_eff %>% 
  filter(Effect == "Direct" | Effect == "Indirect") %>% 
  ggplot(aes(x = Brain, y = estimate, colour = Brain)) +
  #geom_errorbar(aes(ymin=len-ci, ymax=len+ci), colour="black", width=.1, position=pd) +
  geom_errorbar(aes(ymin = conf.low, ymax = conf.high, colour = Brain), 
                 width = .1, alpha = .5) +
  geom_point(fill = "white", shape = 21) +
  labs(y = "Beta Estimates") +
  scale_colour_manual(values = color_1)+
  facet_wrap(~Effect, scale = "free")+
  theme_minimal()+
  theme(axis.title.x = element_blank(), axis.ticks.x = element_blank(), 
        axis.line.x = element_blank(), axis.text.x = element_blank())

4.2 Replication Mediation

We use the same mediation model syntax as specified above.

Create combinations and run models combinations for IV, Mediation or Brain DVs to run 5 mediation. Here, we are running the variables for the replication:

IV: FamEnv_Fact, Meditor: p_puberty, DVs: Amygdala_vol, ACC_CT, ACC_CA, L_AmygCing_rest, R_AmygCing_rest

# list variables for mediation
Rep_iv = c("FamEnv_Fact")
Rep_dv = c("Amygdala_vol","ACC_CT","ACC_CA","L_AmygCing_rest","R_AmygCing_rest")
Rep_m = c("p_puberty")

# Future users, if permuting across covariates, uncomment below & update input/model code above
#Rep_cov = c("Age")

Running the model. After the initial instance of running this model, we save the table output and do not run it next time. WIthout using the parallelized version of the multiverse function written above (which uses lapply), the duration of running this model on a macOS Monterey, 2.3GHz 8-code i9, 16GB ram is 36 minutes.

The code here below used to confirm initial instances of the mediation model, to ensure the models 1) run without major errors, 2) appropriate data is extracted and 3) no egregrious problems arise.

After the initial instance, the R code to run this model is not ran. Instead, after running the model the first time, we save the table with the estimates and load them back into Rstudio to use in subsequent steps. This is to save on machine resources/time and to increase efficiency in knitting the final version of the .html document.

Here we write the extract_mediate function. First, we pull the associated X, Y, M variable names to replace the model values with our variable names for the plots. Using the data pulled by model_parameters, we pull the associated [row,column] from the model_parameters() output for the model. Here we focus on the Direct, Indirect, Total, Apath, and Bpath, extracting the beta, standard error, association upper/lower bound 95% CI, and p-value

4.2.1 Parameters Output

In the function we defined, `mediation_specr’, we model these parameters using lapply. This simply a function that can work across lists that we created – think of it as cleaner version of a for loop. We combined the extracted values across rows (bind_rows), then, we use a combination of gather, separate and spread to get ou data frame in the manner we want.

# load the prev. data
mediation_replicate <- read.csv("./Output/MedReplicate_Table_April_2022_renamed.csv")

# summarize the data

mediation_replicate %>% 
  kbl(digits = 4, booktabs = TRUE) %>% 
  kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive"),
                full_width = F, html_font = "Times")
X Y M Observations N_Parameters Effect conf.high conf.low estimate p.value std.error
FamEnv Fact ACC CA PDS Par 6658 13 Apath -0.1420 -0.1835 -0.1627 0.0000 0.0106
FamEnv Fact ACC CA PDS Par 6658 13 Bpath 0.0659 0.0078 0.0369 0.0128 0.0148
FamEnv Fact ACC CA PDS Par 6658 13 Direct 0.0208 -0.0287 -0.0040 0.7535 0.0126
FamEnv Fact ACC CA PDS Par 6658 13 Indirect -0.0012 -0.0108 -0.0060 0.0140 0.0024
FamEnv Fact ACC CA PDS Par 6658 13 Total 0.0144 -0.0343 -0.0100 0.4222 0.0124
FamEnv Fact ACC CT PDS Par 6658 13 Apath -0.1420 -0.1832 -0.1626 0.0000 0.0105
FamEnv Fact ACC CT PDS Par 6658 13 Bpath -0.0188 -0.0787 -0.0487 0.0014 0.0153
FamEnv Fact ACC CT PDS Par 6658 13 Direct 0.0066 -0.0449 -0.0191 0.1458 0.0132
FamEnv Fact ACC CT PDS Par 6658 13 Indirect 0.0129 0.0030 0.0079 0.0016 0.0025
FamEnv Fact ACC CT PDS Par 6658 13 Total 0.0140 -0.0364 -0.0112 0.3837 0.0129
FamEnv Fact Amygdala vol PDS Par 6658 13 Apath -0.1422 -0.1833 -0.1627 0.0000 0.0105
FamEnv Fact Amygdala vol PDS Par 6658 13 Bpath 0.0544 -0.0036 0.0254 0.0857 0.0148
FamEnv Fact Amygdala vol PDS Par 6658 13 Direct 0.0239 -0.0255 -0.0008 0.9497 0.0126
FamEnv Fact Amygdala vol PDS Par 6658 13 Indirect 0.0006 -0.0089 -0.0041 0.0872 0.0024
FamEnv Fact Amygdala vol PDS Par 6658 13 Total 0.0193 -0.0292 -0.0049 0.6898 0.0124
FamEnv Fact L AmygCON PDS Par 6658 13 Apath -0.1421 -0.1839 -0.1630 0.0000 0.0106
FamEnv Fact L AmygCON PDS Par 6658 13 Bpath -0.0414 -0.1018 -0.0716 0.0000 0.0154
FamEnv Fact L AmygCON PDS Par 6658 13 Direct 0.1210 0.0731 0.0971 0.0000 0.0122
FamEnv Fact L AmygCON PDS Par 6658 13 Indirect 0.0168 0.0065 0.0117 0.0000 0.0026
FamEnv Fact L AmygCON PDS Par 6658 13 Total 0.1323 0.0852 0.1088 0.0000 0.0120
FamEnv Fact R AmygCON PDS Par 6658 13 Apath -0.1422 -0.1834 -0.1628 0.0000 0.0105
FamEnv Fact R AmygCON PDS Par 6658 13 Bpath -0.0356 -0.0942 -0.0649 0.0000 0.0149
FamEnv Fact R AmygCON PDS Par 6658 13 Direct 0.0995 0.0502 0.0748 0.0000 0.0126
FamEnv Fact R AmygCON PDS Par 6658 13 Indirect 0.0156 0.0056 0.0106 0.0000 0.0025
FamEnv Fact R AmygCON PDS Par 6658 13 Total 0.1097 0.0611 0.0854 0.0000 0.0124

4.3 Compare: Orig. v Replication

First, we pull Brain variable and indirect/direct/total/apath/bpath betas and associated p.values from our replication model for the 5 DVs

replication_to_compare <- mediation_replicate %>% 
  dplyr::select(Y, Effect, estimate, p.value, conf.high, conf.low) %>% 
  rename("Brain" = Y)

For the conceptual replication, we proposed to check several things:

  1. Whether the estimates in the replication study are in the same direction as the original
  2. Whether the estimates in the replication study are significant as in the original
  3. Whether the estimates in the replication study overlap with the 95% CI from the original study for that estimate. For this, we do the reverse, too, plot and obverse whether the original betas overlation with the replication betas +/- 95% CI.

4.3.1 Plotting Original/Replicated Estimates

First, we provide a visual representation of item 1 - 3 noted above. We overlay the original study beta estimates + 95% CI and the replication model beta estimates for each mediation path across [5] brain DVs mediation models. The original beta estimate is a colored circle (differentiated by path) and the replicated estimate is a colored ‘X’ associated with a similar color for each mediation path and brain DV.

We focus on the 95% CI due to the fact that across all random samples for the size of this data, 95% of those samples will have a 95% confidence interval that will contain the estimated value that represents the data with similar characteristics. In other words, if we sampled 100 participants, got our means and SD and we were to repeatedly sample based on the sample size, mean, and SD we would get an estimate with a lower and upper bound, or our 95% CI.

color_1 <- cbPalette <- c("black", "chocolate1", "firebrick", "slateblue", "turquoise4")# six colours for six factors

#jpeg("Figures/Aim1/ReplicBeta_v_Orig95ci.jpeg", units = "in",
#     width = 8, height = 5, res = 300)

Thij_eff %>% 
  #filter(!Effect == "Total") %>% 
  ggplot(aes(x = Brain, y = estimate, colour = Brain)) +
  #geom_errorbar(aes(ymin=len-ci, ymax=len+ci), colour="black", width=.1, position=pd) +
  geom_errorbar(aes(ymin = conf.low, ymax = conf.high, colour = Brain), 
                 width = .1, alpha = .5) +
  geom_point(fill = "white", shape = 21) +
  geom_point(data = replication_to_compare, aes(x = Brain, y = estimate, colour = Brain), 
             shape = 4, position = "jitter") +
  labs(y = "Beta Estimates", 
       colour = "Brain Outcome"
       #caption = "o = β From Thijssen et al. \n x: β From Replication"
       ) +
  scale_colour_manual(values = color_1)+
  facet_wrap(~Effect, scale = "free")+
  theme_minimal()+
  theme(axis.title.x = element_blank(), axis.ticks.x = element_blank(), 
        axis.line.x = element_blank(), axis.text.x = element_blank())

#dev.off()

sub_effects_a <- Thij_eff %>% 
  filter(Effect == "Direct" | Effect == "Indirect") %>% 
  ggplot(aes(x = Brain, y = estimate, colour = Brain)) +
  #geom_errorbar(aes(ymin=len-ci, ymax=len+ci), colour="black", width=.1, position=pd) +
  geom_errorbar(aes(ymin = conf.low, ymax = conf.high, colour = Brain), 
                 width = .1, alpha = .5) +
  geom_point(fill = "white", shape = 21) +
  geom_point(data = replication_to_compare %>% filter(Effect == "Direct" | Effect == "Indirect"), aes(x = Brain, y = estimate, colour = Brain), 
             shape = 4, position = "jitter") +
  labs(y = "Beta Estimates", 
       colour = "Brain Outcome"
       #caption = "o = β From Thijssen et al. \n x: β From Replication"
       ) +
  scale_colour_manual(values = color_1)+
  facet_wrap(~Effect, scale = "free")+
  theme_minimal()+
  theme(axis.title.x = element_blank(), axis.ticks.x = element_blank(), 
        axis.line.x = element_blank(), axis.text.x = element_blank(),
        legend.position = "none")

Here, we plot the inverse of the above. Above we visualized the beta estimates & 95% CI from the original study and how the replicated beta estimates overlapped with this. Here, we visualize the replicated beta estimates & their associated 95% and see how the original beta estimates replicate with them.

color_1 <- cbPalette <- c("black", "chocolate1", "firebrick", "slateblue", "turquoise4")# six colours for six factors

#jpeg("Figures/Aim1/OrigBeta_v_Repl95ci.jpeg", units = "in",
#     width = 8, height = 5, res = 300)

replication_to_compare %>% 
  #filter(!Effect == "Total") %>% 
  ggplot(aes(x = Brain, y = estimate, colour = Brain)) +
  #geom_errorbar(aes(ymin=len-ci, ymax=len+ci), colour="black", width=.1, position=pd) +
  geom_errorbar(aes(ymin = conf.low, ymax = conf.high, colour = Brain), 
                 width = .1, alpha = .5) +
  geom_point(fill = "white", shape = 4) +
  geom_point(data = Thij_eff, aes(x = Brain, y = estimate, colour = Brain), 
             shape = 21, position = "jitter") +
  labs(y = "Beta Estimates", 
       colour = "Brain Outcome"
  #     caption = "o = β From Thijssen et al. \n x: β From Replication"
       ) +
  scale_colour_manual(values = color_1)+
  facet_wrap(~Effect, scale = "free")+
  theme_minimal()+
  theme(axis.title.x = element_blank(), axis.ticks.x = element_blank(), 
        axis.line.x = element_blank(), axis.text.x = element_blank())

#dev.off()


sub_effects_b <- replication_to_compare %>% 
  filter(Effect == "Direct" | Effect == "Indirect") %>% 
  ggplot(aes(x = Brain, y = estimate, colour = Brain)) +
  #geom_errorbar(aes(ymin=len-ci, ymax=len+ci), colour="black", width=.1, position=pd) +
  geom_errorbar(aes(ymin = conf.low, ymax = conf.high, colour = Brain), 
                 width = .1, alpha = .5) +
  geom_point(fill = "white", shape = 4) +
  geom_point(data = Thij_eff %>% filter(Effect == "Direct" | Effect == "Indirect"), 
             aes(x = Brain, y = estimate, colour = Brain), 
             shape = 21, position = "jitter") +
  labs(y = "", 
       colour = "Brain Outcome") +
  scale_colour_manual(values = color_1)+
  facet_wrap(~Effect, scale = "free")+
  theme_minimal()+
  theme(axis.title.x = element_blank(), axis.ticks.x = element_blank(), 
        axis.line.x = element_blank(), axis.text.x = element_blank())


#jpeg("Figures/Aim1/Combined_DirInd_plt.jpeg", units = "in",
#     width = 8, height = 5, res = 300)

#sub_effects_a / sub_effects_b + plot_layout(guides = 'collect')

cowplot::plot_grid(sub_effects_a, sub_effects_b, ncol = 1, align = "v", axis = 'tblr',
                       labels = c('A', 'B'), rel_heights = c(1, 1),
                   label_fontfamily = "Times", label_size = 14)

#dev.off()

To understand the confidence intervals a bit better, here we plot the distributions of the 5 brain outcomes. This may help inform why the confidence intervals are wider for some brain, such as ACC CA/CT and volumen, but not others, such as the L/R amygdala & CING connectivity. We plot this in two different ways, where the y-scale is fixed across all of the variables and where the y-scale is free to vary across the brain types.

data %>% 
  pivot_longer(c(Amygdala_vol,ACC_CT,ACC_CA,
                 L_AmygCing_rest,R_AmygCing_rest), 
  names_to = "Brain_ROI", values_to = "Values") %>% 
  ggplot(aes(x = Values)) +
  geom_density() +
  facet_wrap(vars(Brain_ROI), ncol = 3) +
  labs(title = "Fixed scale", x = "Brain ROI", y = "Value Distributions", 
       caption = "vol = volume; ACC = Anterior Cingulate Cortex; CT = cortical thickness; /n 
       CA = cortical area; L = Left, R = Right; AmygCing_rest = Amygdala-CON rsfMRI") +
  theme_minimal()

data %>% 
  pivot_longer(c(Amygdala_vol,ACC_CT,ACC_CA,
                 L_AmygCing_rest,R_AmygCing_rest), 
  names_to = "Brain_ROI", values_to = "Values") %>% 
  ggplot(aes(x = Values)) +
  geom_density() +
  facet_wrap(vars(Brain_ROI), ncol = 3, scales = "free") +
  labs(title = "Free scale", x = "Brain ROI", y = "Value Distributions", 
       caption = "vol = volume; ACC = Anterior Cingulate Cortex; CT = cortical thickness; /n 
       CA = cortical area; L = Left, R = Right; AmygCing_rest = Amygdala-CON rsfMRI") +
  theme_minimal()

4.3.2 Calculating overlap btwn Original/Replicated Estimates

Here, we generate some estimates to determine p-value (i.e., p < .001, < .01, < .05) threshold for each estimate using case_when. Then, we ask a True (1)/False(0) statement using if_else within mutate to determine whether 1) the sign of the effect (+/-) is in the same direction using the sign function, 2) the effect from replicated data found between 95% CI in original study and 2) that the original p-value category (i.e., p > 05 or p < .05) is similar across the two. Before we do this, we combine the original data output (dataframe(Thij_eff) and the replicated data (data.frame(replication_to_compare)) by the “brain” and the “effect” overlap. We provide the suffix “_OG” for the original coefficients and “_Rep” for the replication.

combined_OG_Rep <-left_join(Thij_eff, replication_to_compare, 
          by = c("Brain","Effect"),
          suffix = c("_OG","_Rep"))

combined_OG_Rep <- combined_OG_Rep %>% 
  mutate(pval_Rep = case_when(p.value_Rep < .05 ~ "< .05",
                               TRUE ~ "> .05")) %>% 
  mutate(RepB_BtwnOrig_95CI = 
           if_else(estimate_Rep < conf.high_OG & conf.low_OG < estimate_Rep,1,0), 
         OrigB_BtwnRep_95CI = 
           if_else(estimate_OG < conf.high_Rep & conf.low_Rep < estimate_OG,1,0),
         same_p_cat = if_else(p.value_OG == pval_Rep,1,0),
         same_sign = if_else(sign(estimate_Rep) == sign(estimate_OG), 1, 0)
  )

write.csv(x = combined_OG_Rep, file = "./Output/Effect_OGRep_Overlap.csv", row.names = F)

Now that we have generated our variables, we can quantify the % of which overlap by using the mean() within the summarize() wrap. Given that our values are 1/0, using the mean works, given that we add/divide.

For all parameter estimates (total, direct, indirect, a/b path)

combined_OG_Rep %>% 
  dplyr::summarize('RepBeta_btwn_95CI_OfOrig' = mean(RepB_BtwnOrig_95CI, na.rm = T),
            'OrigBeta_btwn_95CI_OfRep' = mean(OrigB_BtwnRep_95CI, na.rm = T),
            'RepP_SameCategory_AsOrig' = mean(same_p_cat, na.rm = T),
            'RepBetaSign_SameDir_AsOrig' = mean(same_sign, na.rm = T),)
##   RepBeta_btwn_95CI_OfOrig OrigBeta_btwn_95CI_OfRep RepP_SameCategory_AsOrig
## 1                     0.92                     0.48                     0.68
##   RepBetaSign_SameDir_AsOrig
## 1                       0.84
#jpeg("Figures/Aim1/RepOrigOverlap_Barplot.jpeg", units = "in",
#     width = 10, height = 5, res = 300)

combined_OG_Rep %>% 
  dplyr::summarize('RepBeta_btwn_95CI_OfOrig' = mean(RepB_BtwnOrig_95CI, na.rm = T),
            'OrigBeta_btwn_95CI_OfRep' = mean(OrigB_BtwnRep_95CI, na.rm = T),
            'RepP_SameCategory_AsOrig' = mean(same_p_cat, na.rm = T),
            'RepBetaSign_SameDir_AsOrig' = mean(same_sign, na.rm = T)) %>% 
  gather(., key = X, value = Y, RepBeta_btwn_95CI_OfOrig,OrigBeta_btwn_95CI_OfRep,
         RepP_SameCategory_AsOrig, RepBetaSign_SameDir_AsOrig) %>% 
  ggplot(aes(x = X, y = Y, fill = X)) +
  geom_bar(stat = "identity") +
  labs(col = "Replicated Estimates")+
  ylim(c(0.0,1.0))+
  ylab("Percentage") +
  xlab("")+
  theme_minimal() +
  theme(axis.text.x = element_blank(), legend.title = element_blank(), 
        legend.position = "bottom")+
  scale_fill_manual(values=c("#999999", "coral2", "#E69F00", "#56B4E9"),
                    labels = c(
                               "Original β Between \n95% CI for Replication β",
                               "Replication β Between \n95% CI around Original β",
                               "Replication β Sign Same \nDirection As Original β",
                               "Replication p-val Same \nCategory as Original p-val"))

#dev.off()

For Indirect and Direct paths only, as preregistered for Aim 1

combined_OG_Rep %>% 
  filter(Effect == "Direct" | Effect == "Indirect") %>%
  dplyr::summarize('RepBeta_btwn_95CI_OfOrig' = mean(RepB_BtwnOrig_95CI, na.rm = T),
            'OrigBeta_btwn_95CI_OfRep' = mean(OrigB_BtwnRep_95CI, na.rm = T),
            'RepP_SameCategory_AsOrig' = mean(same_p_cat, na.rm = T),
            'RepBetaSign_SameDir_AsOrig' = mean(same_sign, na.rm = T),)
##   RepBeta_btwn_95CI_OfOrig OrigBeta_btwn_95CI_OfRep RepP_SameCategory_AsOrig
## 1                        1                      0.5                      0.6
##   RepBetaSign_SameDir_AsOrig
## 1                        0.9
#jpeg("Figures/Aim1/RepOrigOverlap_Barplot_DirIndOnly.jpeg", units = "in",
 #    width = 10, height = 5, res = 300)

combined_OG_Rep %>% 
  filter(Effect == "Direct" | Effect == "Indirect") %>% 
  dplyr::summarize('RepBeta_btwn_95CI_OfOrig' = mean(RepB_BtwnOrig_95CI, na.rm = T),
            'OrigBeta_btwn_95CI_OfRep' = mean(OrigB_BtwnRep_95CI, na.rm = T),
            'RepP_SameCategory_AsOrig' = mean(same_p_cat, na.rm = T),
            'RepBetaSign_SameDir_AsOrig' = mean(same_sign, na.rm = T)) %>% 
  gather(., key = X, value = Y, RepBeta_btwn_95CI_OfOrig,OrigBeta_btwn_95CI_OfRep,
         RepP_SameCategory_AsOrig, RepBetaSign_SameDir_AsOrig) %>% 
  ggplot(aes(x = X, y = Y, fill = X)) +
  geom_bar(stat = "identity") +
  labs(col = "Replicated Estimates")+
  ylim(c(0.0,1.0))+
  ylab("Percentage") +
  xlab("")+
  theme_minimal() +
  theme(axis.text.x = element_blank(), legend.title = element_blank(), 
        legend.position = "bottom")+
  scale_fill_manual(values=c("#999999", "coral2", "#E69F00", "#56B4E9"),
                    labels = c(
                               "Original β Between \n95% CI for Replication β",
                               "Replication β Between \n95% CI around Original β",
                               "Replication β Sign Same \nDirection As Original β",
                               "Replication p-val Same \nCategory as Original p-val"))

#dev.off()

4.4 Subjective Eval Replication

Here we randomly select 5 co-authors out of a list of 8 to subjectively evaluate the replication of the effects. As preregistered, we do not include original authors in this list.

First, we randomly generate a seed between 1 and 1000, and then randomly sample 5 names from a list of 8, without replacement. We randomize the seed so the resulting values are truly random. Otherwise, by having a list of 8 values we would know prospectively which 8 would be selected based for a set.seed of say, 100.

Note, after the initial run, to avoid bias this code is not rerun. Thus, a single file is generated. The initial code was ran one day after the preregistration: February 19th, 2022, 9:45AM PST.

4.5 Plot Subjective Replication

subj_tally <- read.csv("./Output/Replication_Tally_plt.csv")

colors <- cbPalette <- c("black", "chocolate1", "firebrick", "slateblue", "turquoise4")

#jpeg("Figures/Aim1/Replication_Agreement.jpeg", units = "in",
#     width = 10, height = 5, res = 300)
subj_tally %>% 
  ggplot(aes(x = Brain, y = Overall_Percent, fill = Brain)) +
  geom_bar(stat = "identity") +
  labs(col = "Replicated Estimates", caption = "Dotted, Dashed line = 50% Agreement")+
  geom_hline(yintercept = .50, linetype = "dotdash", colour = "grey")+ 
  ylim(c(0.0,1.0))+
  ylab("Percentage Agreement (%)") +
  xlab("")+
  facet_wrap(~Effect)+
  scale_fill_manual(values = colors)+
  theme_minimal() +
  theme(axis.text.x = element_blank(), legend.title = element_blank(), 
        legend.position = "bottom")

#dev.off()

5 Aim 2

5.1 Creating list of variables

Here we create a list of variables that we will be using. For example, we are running models across different combinations of IV, DV, and Mediators. We add these to a list.

In our function, we use exapnd.grid to create all permutations for our “X”, “Y” and “M” columns that’ll be used in the mediation model. We save this as as a matrix.

Recall, we have reverse coded variables in prior R steps (such as FES youth, FES Parent, Par Factor in Part2 & 3) to maintain consistency in the interpretation. In otherwords, higher scores on the variable reflect more positive (i.e., less stressful) environment.

# list variables for mediation
IVs = c("FamEnv_Fact","Demo_Fact","Child_Fact","Par_Fact",
       "FES_youth","FES_parent", 
       "PMON","CRPBI","Avg_IncomeEduc")

M = c("p_puberty","y_puberty","avg_puberty")

DVs = c("Amygdala_vol","ACC_CT","ACC_CA","L_AmygCing_rest","R_AmygCing_rest")

5.2 Mediation model

In the real models we will fit a mediation model degenerated from Steps 1 - 3. This, of course, would be expanded by including the brain variables. The goal would be to create a loop/function that will run the mediation with each version of:

Family

  • Overall factor (Family Environment).
  • Sub factor A (Child)
  • Sub factor B (Parent)
  • Sub factor C (Demographic)
  • Scale level (e.g., monitoring, acceptance, etc)

Puberty

  • Child SR
  • Parent SR

Brain

  • Using similar brain outcomes as original paper, using 5 of the 6 brain outcomes, given DWI sig. preprocessing changes between release 1 & 2

While for the real analyses we will use the variables from the original data, here we will focus on our simulated variables. So the mediation model will be comprised using these fake variables.

5.3 Run mediation models

Here we take the permutations of our X, Y, M variables, combined it with apply() to run a quick function that will run our mediation models using lavaan. For specifics related to this function, revisit Multiverse Function above

Running the multivers model. After the initial instance of running this model, we save the table output and do not run it subsequent times R is restarted. Without using the parallelized version of the multiverse function written above (which uses lapply), the duration of running this model on a macOS Monterey, 2.3GHz 8-code i9, 16GB ram is ~ 16hrs.

5.4 Estimates

here we extract and review the variables in the model + standardized parameters, standard error, 95% CI (lb/ub) and p-value.

Below we can visualize this complete set out coefficients for our models. This output is associated with the length N = 135 models that we were intending to get. This is product for X, M, Y.

# Load in multiverse tables
mediation_multiverse <- read.csv("./Output/Multiverse_Table_July_2022_renamed.csv") 

mediation_multiverse %>% 
  kbl(digits = 4, booktabs = TRUE) %>% 
  kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive"),
                full_width = F, html_font = "Times")
X Y M Observations N_Parameters Effect conf.high conf.low estimate p.value std.error
Avg IncEdu ACC CA PDS ParYth 6063 13 Apath -0.1647 -0.2170 -0.1909 0.0000 0.0133
Avg IncEdu ACC CA PDS ParYth 6063 13 Bpath 0.0445 -0.0188 0.0129 0.4257 0.0161
Avg IncEdu ACC CA PDS ParYth 6063 13 Direct -0.0158 -0.0688 -0.0423 0.0018 0.0135
Avg IncEdu ACC CA PDS ParYth 6063 13 Indirect 0.0036 -0.0085 -0.0025 0.4268 0.0031
Avg IncEdu ACC CA PDS ParYth 6063 13 Total -0.0188 -0.0706 -0.0447 0.0007 0.0132
Avg IncEdu ACC CA PDS Par 6063 13 Apath -0.1561 -0.2041 -0.1801 0.0000 0.0122
Avg IncEdu ACC CA PDS Par 6063 13 Bpath 0.0650 0.0042 0.0346 0.0259 0.0155
Avg IncEdu ACC CA PDS Par 6063 13 Direct -0.0114 -0.0656 -0.0385 0.0054 0.0138
Avg IncEdu ACC CA PDS Par 6063 13 Indirect -0.0007 -0.0118 -0.0062 0.0278 0.0028
Avg IncEdu ACC CA PDS Par 6063 13 Total -0.0182 -0.0712 -0.0447 0.0010 0.0135
Avg IncEdu ACC CA PDS Yth 6063 13 Apath -0.1179 -0.1749 -0.1464 0.0000 0.0145
Avg IncEdu ACC CA PDS Yth 6063 13 Bpath 0.0271 -0.0304 -0.0016 0.9121 0.0147
Avg IncEdu ACC CA PDS Yth 6063 13 Direct -0.0181 -0.0719 -0.0450 0.0011 0.0137
Avg IncEdu ACC CA PDS Yth 6063 13 Indirect 0.0044 -0.0040 0.0002 0.9121 0.0021
Avg IncEdu ACC CA PDS Yth 6063 13 Total -0.0182 -0.0713 -0.0447 0.0010 0.0135
Avg IncEdu ACC CT PDS ParYth 6063 13 Apath -0.1645 -0.2172 -0.1909 0.0000 0.0134
Avg IncEdu ACC CT PDS ParYth 6063 13 Bpath 0.0032 -0.0617 -0.0292 0.0775 0.0166
Avg IncEdu ACC CT PDS ParYth 6063 13 Direct 0.0425 -0.0120 0.0152 0.2727 0.0139
Avg IncEdu ACC CT PDS ParYth 6063 13 Indirect 0.0118 -0.0007 0.0056 0.0803 0.0032
Avg IncEdu ACC CT PDS ParYth 6063 13 Total 0.0473 -0.0056 0.0208 0.1226 0.0135
Avg IncEdu ACC CT PDS Par 6063 13 Apath -0.1567 -0.2033 -0.1800 0.0000 0.0119
Avg IncEdu ACC CT PDS Par 6063 13 Bpath -0.0162 -0.0789 -0.0476 0.0030 0.0160
Avg IncEdu ACC CT PDS Par 6063 13 Direct 0.0391 -0.0146 0.0122 0.3728 0.0137
Avg IncEdu ACC CT PDS Par 6063 13 Indirect 0.0143 0.0028 0.0086 0.0035 0.0029
Avg IncEdu ACC CT PDS Par 6063 13 Total 0.0469 -0.0053 0.0208 0.1186 0.0133
Avg IncEdu ACC CT PDS Yth 6063 13 Apath -0.1177 -0.1752 -0.1464 0.0000 0.0147
Avg IncEdu ACC CT PDS Yth 6063 13 Bpath 0.0232 -0.0343 -0.0055 0.7057 0.0147
Avg IncEdu ACC CT PDS Yth 6063 13 Direct 0.0474 -0.0074 0.0200 0.1525 0.0140
Avg IncEdu ACC CT PDS Yth 6063 13 Indirect 0.0050 -0.0034 0.0008 0.7060 0.0022
Avg IncEdu ACC CT PDS Yth 6063 13 Total 0.0478 -0.0061 0.0208 0.1302 0.0138
Avg IncEdu Amygdala vol PDS ParYth 6063 13 Apath -0.1645 -0.2173 -0.1909 0.0000 0.0135
Avg IncEdu Amygdala vol PDS ParYth 6063 13 Bpath 0.0598 -0.0043 0.0277 0.0895 0.0163
Avg IncEdu Amygdala vol PDS ParYth 6063 13 Direct 0.0321 -0.0209 0.0056 0.6784 0.0135
Avg IncEdu Amygdala vol PDS ParYth 6063 13 Indirect 0.0009 -0.0114 -0.0053 0.0916 0.0031
Avg IncEdu Amygdala vol PDS ParYth 6063 13 Total 0.0261 -0.0255 0.0003 0.9816 0.0132
Avg IncEdu Amygdala vol PDS Par 6063 13 Apath -0.1560 -0.2042 -0.1801 0.0000 0.0123
Avg IncEdu Amygdala vol PDS Par 6063 13 Bpath 0.0563 -0.0039 0.0262 0.0875 0.0153
Avg IncEdu Amygdala vol PDS Par 6063 13 Direct 0.0308 -0.0207 0.0051 0.7010 0.0132
Avg IncEdu Amygdala vol PDS Par 6063 13 Indirect 0.0007 -0.0102 -0.0047 0.0893 0.0028
Avg IncEdu Amygdala vol PDS Par 6063 13 Total 0.0257 -0.0250 0.0003 0.9796 0.0129
Avg IncEdu Amygdala vol PDS Yth 6063 13 Apath -0.1176 -0.1754 -0.1465 0.0000 0.0148
Avg IncEdu Amygdala vol PDS Yth 6063 13 Bpath 0.0417 -0.0159 0.0129 0.3793 0.0147
Avg IncEdu Amygdala vol PDS Yth 6063 13 Direct 0.0278 -0.0234 0.0022 0.8671 0.0131
Avg IncEdu Amygdala vol PDS Yth 6063 13 Indirect 0.0023 -0.0061 -0.0019 0.3805 0.0022
Avg IncEdu Amygdala vol PDS Yth 6063 13 Total 0.0256 -0.0250 0.0003 0.9819 0.0129
Avg IncEdu L AmygCON PDS ParYth 6063 13 Apath -0.1646 -0.2175 -0.1911 0.0000 0.0135
Avg IncEdu L AmygCON PDS ParYth 6063 13 Bpath -0.0147 -0.0825 -0.0486 0.0050 0.0173
Avg IncEdu L AmygCON PDS ParYth 6063 13 Direct 0.1544 0.0968 0.1256 0.0000 0.0147
Avg IncEdu L AmygCON PDS ParYth 6063 13 Indirect 0.0159 0.0026 0.0093 0.0061 0.0034
Avg IncEdu L AmygCON PDS ParYth 6063 13 Total 0.1627 0.1070 0.1349 0.0000 0.0142
Avg IncEdu L AmygCON PDS Par 6063 13 Apath -0.1566 -0.2040 -0.1803 0.0000 0.0121
Avg IncEdu L AmygCON PDS Par 6063 13 Bpath -0.0404 -0.1027 -0.0716 0.0000 0.0159
Avg IncEdu L AmygCON PDS Par 6063 13 Direct 0.1502 0.0941 0.1222 0.0000 0.0143
Avg IncEdu L AmygCON PDS Par 6063 13 Indirect 0.0188 0.0070 0.0129 0.0000 0.0030
Avg IncEdu L AmygCON PDS Par 6063 13 Total 0.1625 0.1077 0.1351 0.0000 0.0140
Avg IncEdu L AmygCON PDS Yth 6063 13 Apath -0.1181 -0.1748 -0.1465 0.0000 0.0145
Avg IncEdu L AmygCON PDS Yth 6063 13 Bpath 0.0162 -0.0439 -0.0139 0.3651 0.0153
Avg IncEdu L AmygCON PDS Yth 6063 13 Direct 0.1612 0.1045 0.1329 0.0000 0.0145
Avg IncEdu L AmygCON PDS Yth 6063 13 Indirect 0.0065 -0.0024 0.0020 0.3676 0.0023
Avg IncEdu L AmygCON PDS Yth 6063 13 Total 0.1627 0.1071 0.1349 0.0000 0.0142
Avg IncEdu R AmygCON PDS ParYth 6063 13 Apath -0.1650 -0.2166 -0.1908 0.0000 0.0132
Avg IncEdu R AmygCON PDS ParYth 6063 13 Bpath -0.0077 -0.0729 -0.0403 0.0154 0.0166
Avg IncEdu R AmygCON PDS ParYth 6063 13 Direct 0.1353 0.0791 0.1072 0.0000 0.0143
Avg IncEdu R AmygCON PDS ParYth 6063 13 Indirect 0.0140 0.0013 0.0077 0.0175 0.0032
Avg IncEdu R AmygCON PDS ParYth 6063 13 Total 0.1422 0.0876 0.1149 0.0000 0.0139
Avg IncEdu R AmygCON PDS Par 6063 13 Apath -0.1568 -0.2035 -0.1801 0.0000 0.0119
Avg IncEdu R AmygCON PDS Par 6063 13 Bpath -0.0378 -0.0967 -0.0672 0.0000 0.0150
Avg IncEdu R AmygCON PDS Par 6063 13 Direct 0.1308 0.0751 0.1030 0.0000 0.0142
Avg IncEdu R AmygCON PDS Par 6063 13 Indirect 0.0177 0.0066 0.0121 0.0000 0.0028
Avg IncEdu R AmygCON PDS Par 6063 13 Total 0.1426 0.0876 0.1151 0.0000 0.0140
Avg IncEdu R AmygCON PDS Yth 6063 13 Apath -0.1179 -0.1749 -0.1464 0.0000 0.0145
Avg IncEdu R AmygCON PDS Yth 6063 13 Bpath 0.0225 -0.0380 -0.0077 0.6173 0.0154
Avg IncEdu R AmygCON PDS Yth 6063 13 Direct 0.1420 0.0856 0.1138 0.0000 0.0144
Avg IncEdu R AmygCON PDS Yth 6063 13 Indirect 0.0056 -0.0033 0.0011 0.6186 0.0023
Avg IncEdu R AmygCON PDS Yth 6063 13 Total 0.1426 0.0873 0.1149 0.0000 0.0141
Child Fact ACC CA PDS ParYth 6646 13 Apath -0.0731 -0.1227 -0.0979 0.0000 0.0126
Child Fact ACC CA PDS ParYth 6646 13 Bpath 0.0485 -0.0131 0.0177 0.2606 0.0157
Child Fact ACC CA PDS ParYth 6646 13 Direct 0.0613 0.0136 0.0375 0.0021 0.0122
Child Fact ACC CA PDS ParYth 6646 13 Indirect 0.0013 -0.0048 -0.0017 0.2644 0.0016
Child Fact ACC CA PDS ParYth 6646 13 Total 0.0594 0.0121 0.0358 0.0030 0.0121
Child Fact ACC CA PDS Par 6646 13 Apath -0.0267 -0.0694 -0.0480 0.0000 0.0109
Child Fact ACC CA PDS Par 6646 13 Bpath 0.0690 0.0123 0.0407 0.0049 0.0145
Child Fact ACC CA PDS Par 6646 13 Direct 0.0613 0.0141 0.0377 0.0018 0.0121
Child Fact ACC CA PDS Par 6646 13 Indirect -0.0003 -0.0036 -0.0020 0.0170 0.0008
Child Fact ACC CA PDS Par 6646 13 Total 0.0594 0.0121 0.0357 0.0030 0.0121
Child Fact ACC CA PDS Yth 6646 13 Apath -0.0814 -0.1349 -0.1081 0.0000 0.0137
Child Fact ACC CA PDS Yth 6646 13 Bpath 0.0295 -0.0249 0.0023 0.8690 0.0139
Child Fact ACC CA PDS Yth 6646 13 Direct 0.0598 0.0122 0.0360 0.0030 0.0121
Child Fact ACC CA PDS Yth 6646 13 Indirect 0.0027 -0.0032 -0.0002 0.8691 0.0015
Child Fact ACC CA PDS Yth 6646 13 Total 0.0594 0.0121 0.0357 0.0031 0.0121
Child Fact ACC CT PDS ParYth 6646 13 Apath -0.0728 -0.1228 -0.0978 0.0000 0.0128
Child Fact ACC CT PDS ParYth 6646 13 Bpath 0.0035 -0.0581 -0.0273 0.0825 0.0157
Child Fact ACC CT PDS ParYth 6646 13 Direct -0.0095 -0.0601 -0.0348 0.0070 0.0129
Child Fact ACC CT PDS ParYth 6646 13 Indirect 0.0057 -0.0004 0.0027 0.0865 0.0016
Child Fact ACC CT PDS ParYth 6646 13 Total -0.0070 -0.0573 -0.0322 0.0122 0.0128
Child Fact ACC CT PDS Par 6646 13 Apath -0.0270 -0.0690 -0.0480 0.0000 0.0107
Child Fact ACC CT PDS Par 6646 13 Bpath -0.0188 -0.0770 -0.0479 0.0013 0.0148
Child Fact ACC CT PDS Par 6646 13 Direct -0.0094 -0.0595 -0.0344 0.0070 0.0128
Child Fact ACC CT PDS Par 6646 13 Indirect 0.0040 0.0006 0.0023 0.0077 0.0009
Child Fact ACC CT PDS Par 6646 13 Total -0.0071 -0.0572 -0.0321 0.0118 0.0128
Child Fact ACC CT PDS Yth 6646 13 Apath -0.0812 -0.1350 -0.1081 0.0000 0.0137
Child Fact ACC CT PDS Yth 6646 13 Bpath 0.0246 -0.0298 -0.0026 0.8530 0.0139
Child Fact ACC CT PDS Yth 6646 13 Direct -0.0078 -0.0571 -0.0324 0.0100 0.0126
Child Fact ACC CT PDS Yth 6646 13 Indirect 0.0032 -0.0027 0.0003 0.8530 0.0015
Child Fact ACC CT PDS Yth 6646 13 Total -0.0076 -0.0567 -0.0321 0.0102 0.0125
Child Fact Amygdala vol PDS ParYth 6646 13 Apath -0.0725 -0.1232 -0.0979 0.0000 0.0129
Child Fact Amygdala vol PDS ParYth 6646 13 Bpath 0.0593 -0.0026 0.0284 0.0726 0.0158
Child Fact Amygdala vol PDS ParYth 6646 13 Direct 0.0345 -0.0147 0.0099 0.4301 0.0126
Child Fact Amygdala vol PDS ParYth 6646 13 Indirect 0.0004 -0.0059 -0.0028 0.0844 0.0016
Child Fact Amygdala vol PDS ParYth 6646 13 Total 0.0316 -0.0173 0.0071 0.5667 0.0125
Child Fact Amygdala vol PDS Par 6646 13 Apath -0.0262 -0.0697 -0.0479 0.0000 0.0111
Child Fact Amygdala vol PDS Par 6646 13 Bpath 0.0544 -0.0025 0.0259 0.0742 0.0145
Child Fact Amygdala vol PDS Par 6646 13 Direct 0.0330 -0.0163 0.0083 0.5071 0.0126
Child Fact Amygdala vol PDS Par 6646 13 Indirect 0.0003 -0.0027 -0.0012 0.1045 0.0008
Child Fact Amygdala vol PDS Par 6646 13 Total 0.0317 -0.0175 0.0071 0.5717 0.0126
Child Fact Amygdala vol PDS Yth 6646 13 Apath -0.0815 -0.1349 -0.1082 0.0000 0.0136
Child Fact Amygdala vol PDS Yth 6646 13 Bpath 0.0382 -0.0166 0.0108 0.4399 0.0140
Child Fact Amygdala vol PDS Yth 6646 13 Direct 0.0328 -0.0163 0.0083 0.5082 0.0125
Child Fact Amygdala vol PDS Yth 6646 13 Indirect 0.0018 -0.0041 -0.0012 0.4419 0.0015
Child Fact Amygdala vol PDS Yth 6646 13 Total 0.0315 -0.0172 0.0071 0.5669 0.0124
Child Fact L AmygCON PDS ParYth 6646 13 Apath -0.0732 -0.1227 -0.0980 0.0000 0.0126
Child Fact L AmygCON PDS ParYth 6646 13 Bpath -0.0314 -0.0955 -0.0635 0.0001 0.0164
Child Fact L AmygCON PDS ParYth 6646 13 Direct 0.0623 0.0134 0.0378 0.0024 0.0125
Child Fact L AmygCON PDS ParYth 6646 13 Indirect 0.0098 0.0027 0.0062 0.0006 0.0018
Child Fact L AmygCON PDS ParYth 6646 13 Total 0.0681 0.0200 0.0441 0.0003 0.0123
Child Fact L AmygCON PDS Par 6646 13 Apath -0.0259 -0.0701 -0.0480 0.0000 0.0113
Child Fact L AmygCON PDS Par 6646 13 Bpath -0.0590 -0.1177 -0.0883 0.0000 0.0150
Child Fact L AmygCON PDS Par 6646 13 Direct 0.0638 0.0160 0.0399 0.0011 0.0122
Child Fact L AmygCON PDS Par 6646 13 Indirect 0.0066 0.0018 0.0042 0.0005 0.0012
Child Fact L AmygCON PDS Par 6646 13 Total 0.0680 0.0203 0.0441 0.0003 0.0122
Child Fact L AmygCON PDS Yth 6646 13 Apath -0.0815 -0.1348 -0.1081 0.0000 0.0136
Child Fact L AmygCON PDS Yth 6646 13 Bpath 0.0046 -0.0542 -0.0248 0.0987 0.0150
Child Fact L AmygCON PDS Yth 6646 13 Direct 0.0660 0.0166 0.0413 0.0010 0.0126
Child Fact L AmygCON PDS Yth 6646 13 Indirect 0.0059 -0.0006 0.0027 0.1047 0.0017
Child Fact L AmygCON PDS Yth 6646 13 Total 0.0684 0.0195 0.0440 0.0004 0.0125
Child Fact R AmygCON PDS ParYth 6646 13 Apath -0.0733 -0.1228 -0.0980 0.0000 0.0126
Child Fact R AmygCON PDS ParYth 6646 13 Bpath -0.0256 -0.0877 -0.0567 0.0003 0.0158
Child Fact R AmygCON PDS ParYth 6646 13 Direct 0.0614 0.0100 0.0357 0.0065 0.0131
Child Fact R AmygCON PDS ParYth 6646 13 Indirect 0.0089 0.0022 0.0056 0.0013 0.0017
Child Fact R AmygCON PDS ParYth 6646 13 Total 0.0667 0.0159 0.0413 0.0014 0.0130
Child Fact R AmygCON PDS Par 6646 13 Apath -0.0266 -0.0696 -0.0481 0.0000 0.0110
Child Fact R AmygCON PDS Par 6646 13 Bpath -0.0486 -0.1064 -0.0775 0.0000 0.0148
Child Fact R AmygCON PDS Par 6646 13 Direct 0.0625 0.0127 0.0376 0.0030 0.0127
Child Fact R AmygCON PDS Par 6646 13 Indirect 0.0059 0.0015 0.0037 0.0009 0.0011
Child Fact R AmygCON PDS Par 6646 13 Total 0.0662 0.0165 0.0414 0.0011 0.0127
Child Fact R AmygCON PDS Yth 6646 13 Apath -0.0815 -0.1349 -0.1082 0.0000 0.0136
Child Fact R AmygCON PDS Yth 6646 13 Bpath 0.0070 -0.0508 -0.0219 0.1373 0.0148
Child Fact R AmygCON PDS Yth 6646 13 Direct 0.0639 0.0138 0.0388 0.0024 0.0128
Child Fact R AmygCON PDS Yth 6646 13 Indirect 0.0056 -0.0008 0.0024 0.1467 0.0016
Child Fact R AmygCON PDS Yth 6646 13 Total 0.0660 0.0164 0.0412 0.0011 0.0126
Par Accept ACC CA PDS ParYth 6633 13 Apath -0.0469 -0.0968 -0.0718 0.0000 0.0127
Par Accept ACC CA PDS ParYth 6633 13 Bpath 0.0465 -0.0133 0.0166 0.2768 0.0153
Par Accept ACC CA PDS ParYth 6633 13 Direct 0.0617 0.0143 0.0380 0.0017 0.0121
Par Accept ACC CA PDS ParYth 6633 13 Indirect 0.0010 -0.0034 -0.0012 0.2871 0.0011
Par Accept ACC CA PDS ParYth 6633 13 Total 0.0605 0.0132 0.0368 0.0023 0.0121
Par Accept ACC CA PDS Par 6633 13 Apath -0.0209 -0.0638 -0.0423 0.0001 0.0110
Par Accept ACC CA PDS Par 6633 13 Bpath 0.0702 0.0121 0.0412 0.0055 0.0148
Par Accept ACC CA PDS Par 6633 13 Direct 0.0619 0.0153 0.0386 0.0011 0.0119
Par Accept ACC CA PDS Par 6633 13 Indirect -0.0002 -0.0033 -0.0017 0.0252 0.0008
Par Accept ACC CA PDS Par 6633 13 Total 0.0601 0.0136 0.0369 0.0019 0.0119
Par Accept ACC CA PDS Yth 6633 13 Apath -0.0387 -0.0920 -0.0654 0.0000 0.0136
Par Accept ACC CA PDS Yth 6633 13 Bpath 0.0270 -0.0274 -0.0002 0.9892 0.0139
Par Accept ACC CA PDS Yth 6633 13 Direct 0.0606 0.0130 0.0368 0.0024 0.0121
Par Accept ACC CA PDS Yth 6633 13 Indirect 0.0018 -0.0018 0.0000 0.9892 0.0009
Par Accept ACC CA PDS Yth 6633 13 Total 0.0605 0.0131 0.0368 0.0023 0.0121
Par Accept ACC CT PDS ParYth 6633 13 Apath -0.0461 -0.0971 -0.0716 0.0000 0.0130
Par Accept ACC CT PDS ParYth 6633 13 Bpath 0.0083 -0.0539 -0.0228 0.1514 0.0159
Par Accept ACC CT PDS ParYth 6633 13 Direct 0.0094 -0.0403 -0.0155 0.2233 0.0127
Par Accept ACC CT PDS ParYth 6633 13 Indirect 0.0039 -0.0006 0.0016 0.1579 0.0012
Par Accept ACC CT PDS ParYth 6633 13 Total 0.0109 -0.0386 -0.0138 0.2737 0.0126
Par Accept ACC CT PDS Par 6633 13 Apath -0.0207 -0.0638 -0.0422 0.0001 0.0110
Par Accept ACC CT PDS Par 6633 13 Bpath -0.0162 -0.0747 -0.0454 0.0023 0.0149
Par Accept ACC CT PDS Par 6633 13 Direct 0.0089 -0.0405 -0.0158 0.2108 0.0126
Par Accept ACC CT PDS Par 6633 13 Indirect 0.0035 0.0004 0.0019 0.0164 0.0008
Par Accept ACC CT PDS Par 6633 13 Total 0.0108 -0.0385 -0.0139 0.2710 0.0126
Par Accept ACC CT PDS Yth 6633 13 Apath -0.0381 -0.0926 -0.0654 0.0000 0.0139
Par Accept ACC CT PDS Yth 6633 13 Bpath 0.0288 -0.0259 0.0014 0.9174 0.0140
Par Accept ACC CT PDS Yth 6633 13 Direct 0.0112 -0.0386 -0.0137 0.2809 0.0127
Par Accept ACC CT PDS Yth 6633 13 Indirect 0.0017 -0.0019 -0.0001 0.9175 0.0009
Par Accept ACC CT PDS Yth 6633 13 Total 0.0110 -0.0386 -0.0138 0.2761 0.0127
Par Accept Amygdala vol PDS ParYth 6633 13 Apath -0.0475 -0.0959 -0.0717 0.0000 0.0123
Par Accept Amygdala vol PDS ParYth 6633 13 Bpath 0.0563 -0.0047 0.0258 0.0974 0.0156
Par Accept Amygdala vol PDS ParYth 6633 13 Direct 0.0364 -0.0125 0.0119 0.3377 0.0125
Par Accept Amygdala vol PDS ParYth 6633 13 Indirect 0.0005 -0.0042 -0.0018 0.1182 0.0012
Par Accept Amygdala vol PDS ParYth 6633 13 Total 0.0344 -0.0142 0.0101 0.4155 0.0124
Par Accept Amygdala vol PDS Par 6633 13 Apath -0.0207 -0.0637 -0.0422 0.0001 0.0110
Par Accept Amygdala vol PDS Par 6633 13 Bpath 0.0523 -0.0039 0.0242 0.0921 0.0143
Par Accept Amygdala vol PDS Par 6633 13 Direct 0.0356 -0.0133 0.0111 0.3724 0.0125
Par Accept Amygdala vol PDS Par 6633 13 Indirect 0.0003 -0.0023 -0.0010 0.1297 0.0007
Par Accept Amygdala vol PDS Par 6633 13 Total 0.0346 -0.0143 0.0101 0.4178 0.0125
Par Accept Amygdala vol PDS Yth 6633 13 Apath -0.0383 -0.0925 -0.0654 0.0000 0.0138
Par Accept Amygdala vol PDS Yth 6633 13 Bpath 0.0358 -0.0180 0.0089 0.5179 0.0137
Par Accept Amygdala vol PDS Yth 6633 13 Direct 0.0347 -0.0134 0.0107 0.3856 0.0123
Par Accept Amygdala vol PDS Yth 6633 13 Indirect 0.0012 -0.0024 -0.0006 0.5211 0.0009
Par Accept Amygdala vol PDS Yth 6633 13 Total 0.0341 -0.0139 0.0101 0.4107 0.0123
Par Accept L AmygCON PDS ParYth 6633 13 Apath -0.0464 -0.0967 -0.0716 0.0000 0.0128
Par Accept L AmygCON PDS ParYth 6633 13 Bpath -0.0341 -0.0969 -0.0655 0.0000 0.0160
Par Accept L AmygCON PDS ParYth 6633 13 Direct 0.0479 0.0009 0.0244 0.0418 0.0120
Par Accept L AmygCON PDS ParYth 6633 13 Indirect 0.0075 0.0019 0.0047 0.0012 0.0014
Par Accept L AmygCON PDS ParYth 6633 13 Total 0.0524 0.0058 0.0291 0.0144 0.0119
Par Accept L AmygCON PDS Par 6633 13 Apath -0.0207 -0.0634 -0.0420 0.0001 0.0109
Par Accept L AmygCON PDS Par 6633 13 Bpath -0.0613 -0.1203 -0.0908 0.0000 0.0151
Par Accept L AmygCON PDS Par 6633 13 Direct 0.0487 0.0015 0.0251 0.0371 0.0120
Par Accept L AmygCON PDS Par 6633 13 Indirect 0.0061 0.0015 0.0038 0.0012 0.0012
Par Accept L AmygCON PDS Par 6633 13 Total 0.0525 0.0054 0.0289 0.0161 0.0120
Par Accept L AmygCON PDS Yth 6633 13 Apath -0.0381 -0.0926 -0.0654 0.0000 0.0139
Par Accept L AmygCON PDS Yth 6633 13 Bpath 0.0017 -0.0558 -0.0270 0.0654 0.0147
Par Accept L AmygCON PDS Yth 6633 13 Direct 0.0505 0.0043 0.0274 0.0202 0.0118
Par Accept L AmygCON PDS Yth 6633 13 Indirect 0.0038 -0.0003 0.0018 0.0895 0.0010
Par Accept L AmygCON PDS Yth 6633 13 Total 0.0522 0.0062 0.0292 0.0129 0.0117
Par Accept R AmygCON PDS ParYth 6633 13 Apath -0.0469 -0.0966 -0.0717 0.0000 0.0127
Par Accept R AmygCON PDS ParYth 6633 13 Bpath -0.0262 -0.0901 -0.0581 0.0004 0.0163
Par Accept R AmygCON PDS ParYth 6633 13 Direct 0.0477 -0.0015 0.0231 0.0651 0.0126
Par Accept R AmygCON PDS ParYth 6633 13 Indirect 0.0069 0.0015 0.0042 0.0025 0.0014
Par Accept R AmygCON PDS ParYth 6633 13 Total 0.0517 0.0030 0.0273 0.0279 0.0124
Par Accept R AmygCON PDS Par 6633 13 Apath -0.0204 -0.0639 -0.0422 0.0001 0.0111
Par Accept R AmygCON PDS Par 6633 13 Bpath -0.0499 -0.1076 -0.0787 0.0000 0.0147
Par Accept R AmygCON PDS Par 6633 13 Direct 0.0482 -0.0005 0.0239 0.0548 0.0124
Par Accept R AmygCON PDS Par 6633 13 Indirect 0.0054 0.0012 0.0033 0.0020 0.0011
Par Accept R AmygCON PDS Par 6633 13 Total 0.0514 0.0029 0.0272 0.0280 0.0124
Par Accept R AmygCON PDS Yth 6633 13 Apath -0.0382 -0.0926 -0.0654 0.0000 0.0139
Par Accept R AmygCON PDS Yth 6633 13 Bpath 0.0045 -0.0518 -0.0237 0.0991 0.0144
Par Accept R AmygCON PDS Yth 6633 13 Direct 0.0504 0.0014 0.0259 0.0386 0.0125
Par Accept R AmygCON PDS Yth 6633 13 Indirect 0.0035 -0.0004 0.0015 0.1224 0.0010
Par Accept R AmygCON PDS Yth 6633 13 Total 0.0518 0.0030 0.0274 0.0275 0.0124
Demo Fact ACC CA PDS ParYth 6658 13 Apath -0.1998 -0.2487 -0.2242 0.0000 0.0125
Demo Fact ACC CA PDS ParYth 6658 13 Bpath 0.0361 -0.0271 0.0045 0.7817 0.0161
Demo Fact ACC CA PDS ParYth 6658 13 Direct -0.0155 -0.0670 -0.0413 0.0017 0.0131
Demo Fact ACC CA PDS ParYth 6658 13 Indirect 0.0061 -0.0081 -0.0010 0.7817 0.0036
Demo Fact ACC CA PDS ParYth 6658 13 Total -0.0175 -0.0670 -0.0423 0.0008 0.0126
Demo Fact ACC CA PDS Par 6658 13 Apath -0.2009 -0.2441 -0.2225 0.0000 0.0110
Demo Fact ACC CA PDS Par 6658 13 Bpath 0.0577 -0.0016 0.0281 0.0633 0.0151
Demo Fact ACC CA PDS Par 6658 13 Direct -0.0098 -0.0622 -0.0360 0.0070 0.0134
Demo Fact ACC CA PDS Par 6658 13 Indirect 0.0004 -0.0129 -0.0062 0.0645 0.0034
Demo Fact ACC CA PDS Par 6658 13 Total -0.0169 -0.0677 -0.0423 0.0011 0.0130
Demo Fact ACC CA PDS Yth 6658 13 Apath -0.1371 -0.1901 -0.1636 0.0000 0.0135
Demo Fact ACC CA PDS Yth 6658 13 Bpath 0.0200 -0.0350 -0.0075 0.5921 0.0140
Demo Fact ACC CA PDS Yth 6658 13 Direct -0.0175 -0.0695 -0.0435 0.0010 0.0133
Demo Fact ACC CA PDS Yth 6658 13 Indirect 0.0057 -0.0033 0.0012 0.5924 0.0023
Demo Fact ACC CA PDS Yth 6658 13 Total -0.0168 -0.0677 -0.0423 0.0011 0.0130
Demo Fact ACC CT PDS ParYth 6658 13 Apath -0.1999 -0.2485 -0.2242 0.0000 0.0124
Demo Fact ACC CT PDS ParYth 6658 13 Bpath 0.0093 -0.0535 -0.0221 0.1683 0.0160
Demo Fact ACC CT PDS ParYth 6658 13 Direct 0.0298 -0.0239 0.0029 0.8301 0.0137
Demo Fact ACC CT PDS ParYth 6658 13 Indirect 0.0120 -0.0021 0.0050 0.1691 0.0036
Demo Fact ACC CT PDS ParYth 6658 13 Total 0.0336 -0.0178 0.0079 0.5470 0.0131
Demo Fact ACC CT PDS Par 6658 13 Apath -0.2008 -0.2440 -0.2224 0.0000 0.0110
Demo Fact ACC CT PDS Par 6658 13 Bpath -0.0146 -0.0752 -0.0449 0.0037 0.0154
Demo Fact ACC CT PDS Par 6658 13 Direct 0.0245 -0.0287 -0.0021 0.8775 0.0136
Demo Fact ACC CT PDS Par 6658 13 Indirect 0.0168 0.0032 0.0100 0.0041 0.0035
Demo Fact ACC CT PDS Par 6658 13 Total 0.0335 -0.0178 0.0079 0.5466 0.0131
Demo Fact ACC CT PDS Yth 6658 13 Apath -0.1368 -0.1905 -0.1636 0.0000 0.0137
Demo Fact ACC CT PDS Yth 6658 13 Bpath 0.0291 -0.0257 0.0017 0.9045 0.0140
Demo Fact ACC CT PDS Yth 6658 13 Direct 0.0345 -0.0181 0.0082 0.5424 0.0134
Demo Fact ACC CT PDS Yth 6658 13 Indirect 0.0042 -0.0048 -0.0003 0.9045 0.0023
Demo Fact ACC CT PDS Yth 6658 13 Total 0.0336 -0.0178 0.0079 0.5477 0.0131
Demo Fact Amygdala vol PDS ParYth 6658 13 Apath -0.2000 -0.2487 -0.2243 0.0000 0.0124
Demo Fact Amygdala vol PDS ParYth 6658 13 Bpath 0.0618 -0.0013 0.0303 0.0601 0.0161
Demo Fact Amygdala vol PDS ParYth 6658 13 Direct 0.0344 -0.0176 0.0084 0.5282 0.0133
Demo Fact Amygdala vol PDS ParYth 6658 13 Indirect 0.0003 -0.0139 -0.0068 0.0615 0.0036
Demo Fact Amygdala vol PDS ParYth 6658 13 Total 0.0266 -0.0234 0.0016 0.9015 0.0128
Demo Fact Amygdala vol PDS Par 6658 13 Apath -0.2006 -0.2445 -0.2225 0.0000 0.0112
Demo Fact Amygdala vol PDS Par 6658 13 Bpath 0.0571 -0.0008 0.0281 0.0568 0.0148
Demo Fact Amygdala vol PDS Par 6658 13 Direct 0.0332 -0.0175 0.0078 0.5449 0.0129
Demo Fact Amygdala vol PDS Par 6658 13 Indirect 0.0002 -0.0127 -0.0063 0.0577 0.0033
Demo Fact Amygdala vol PDS Par 6658 13 Total 0.0263 -0.0232 0.0016 0.9008 0.0126
Demo Fact Amygdala vol PDS Yth 6658 13 Apath -0.1370 -0.1905 -0.1637 0.0000 0.0136
Demo Fact Amygdala vol PDS Yth 6658 13 Bpath 0.0379 -0.0172 0.0104 0.4613 0.0141
Demo Fact Amygdala vol PDS Yth 6658 13 Direct 0.0285 -0.0220 0.0033 0.8000 0.0129
Demo Fact Amygdala vol PDS Yth 6658 13 Indirect 0.0028 -0.0062 -0.0017 0.4613 0.0023
Demo Fact Amygdala vol PDS Yth 6658 13 Total 0.0263 -0.0232 0.0016 0.9015 0.0126
Demo Fact L AmygCON PDS ParYth 6658 13 Apath -0.1994 -0.2493 -0.2244 0.0000 0.0127
Demo Fact L AmygCON PDS ParYth 6658 13 Bpath -0.0007 -0.0655 -0.0331 0.0455 0.0165
Demo Fact L AmygCON PDS ParYth 6658 13 Direct 0.1658 0.1113 0.1385 0.0000 0.0139
Demo Fact L AmygCON PDS ParYth 6658 13 Indirect 0.0148 0.0000 0.0074 0.0487 0.0038
Demo Fact L AmygCON PDS ParYth 6658 13 Total 0.1719 0.1200 0.1460 0.0000 0.0132
Demo Fact L AmygCON PDS Par 6658 13 Apath -0.2013 -0.2441 -0.2227 0.0000 0.0109
Demo Fact L AmygCON PDS Par 6658 13 Bpath -0.0254 -0.0859 -0.0556 0.0003 0.0154
Demo Fact L AmygCON PDS Par 6658 13 Direct 0.1601 0.1074 0.1338 0.0000 0.0135
Demo Fact L AmygCON PDS Par 6658 13 Indirect 0.0193 0.0055 0.0124 0.0004 0.0035
Demo Fact L AmygCON PDS Par 6658 13 Total 0.1716 0.1207 0.1462 0.0000 0.0130
Demo Fact L AmygCON PDS Yth 6658 13 Apath -0.1373 -0.1900 -0.1636 0.0000 0.0135
Demo Fact L AmygCON PDS Yth 6658 13 Bpath 0.0232 -0.0360 -0.0064 0.6719 0.0151
Demo Fact L AmygCON PDS Yth 6658 13 Direct 0.1709 0.1189 0.1449 0.0000 0.0133
Demo Fact L AmygCON PDS Yth 6658 13 Indirect 0.0059 -0.0038 0.0010 0.6724 0.0025
Demo Fact L AmygCON PDS Yth 6658 13 Total 0.1713 0.1207 0.1460 0.0000 0.0129
Demo Fact R AmygCON PDS ParYth 6658 13 Apath -0.1999 -0.2487 -0.2243 0.0000 0.0125
Demo Fact R AmygCON PDS ParYth 6658 13 Bpath 0.0001 -0.0630 -0.0315 0.0505 0.0161
Demo Fact R AmygCON PDS ParYth 6658 13 Direct 0.1389 0.0850 0.1120 0.0000 0.0138
Demo Fact R AmygCON PDS ParYth 6658 13 Indirect 0.0142 -0.0001 0.0071 0.0526 0.0036
Demo Fact R AmygCON PDS ParYth 6658 13 Total 0.1449 0.0931 0.1190 0.0000 0.0132
Demo Fact R AmygCON PDS Par 6658 13 Apath -0.2010 -0.2442 -0.2226 0.0000 0.0110
Demo Fact R AmygCON PDS Par 6658 13 Bpath -0.0218 -0.0806 -0.0512 0.0007 0.0150
Demo Fact R AmygCON PDS Par 6658 13 Direct 0.1345 0.0811 0.1078 0.0000 0.0136
Demo Fact R AmygCON PDS Par 6658 13 Indirect 0.0180 0.0048 0.0114 0.0008 0.0034
Demo Fact R AmygCON PDS Par 6658 13 Total 0.1450 0.0934 0.1192 0.0000 0.0132
Demo Fact R AmygCON PDS Yth 6658 13 Apath -0.1373 -0.1900 -0.1636 0.0000 0.0134
Demo Fact R AmygCON PDS Yth 6658 13 Bpath 0.0221 -0.0354 -0.0067 0.6485 0.0147
Demo Fact R AmygCON PDS Yth 6658 13 Direct 0.1441 0.0918 0.1179 0.0000 0.0133
Demo Fact R AmygCON PDS Yth 6658 13 Indirect 0.0058 -0.0036 0.0011 0.6489 0.0024
Demo Fact R AmygCON PDS Yth 6658 13 Total 0.1444 0.0936 0.1190 0.0000 0.0130
FamEnv Fact ACC CA PDS ParYth 6658 13 Apath -0.1600 -0.2069 -0.1835 0.0000 0.0120
FamEnv Fact ACC CA PDS ParYth 6658 13 Bpath 0.0417 -0.0196 0.0110 0.4803 0.0156
FamEnv Fact ACC CA PDS ParYth 6658 13 Direct 0.0172 -0.0331 -0.0079 0.5379 0.0128
FamEnv Fact ACC CA PDS ParYth 6658 13 Indirect 0.0036 -0.0077 -0.0020 0.4807 0.0029
FamEnv Fact ACC CA PDS ParYth 6658 13 Total 0.0144 -0.0343 -0.0099 0.4239 0.0124
FamEnv Fact ACC CA PDS Par 6658 13 Apath -0.1421 -0.1834 -0.1627 0.0000 0.0105
FamEnv Fact ACC CA PDS Par 6658 13 Bpath 0.0663 0.0074 0.0369 0.0142 0.0150
FamEnv Fact ACC CA PDS Par 6658 13 Direct 0.0209 -0.0288 -0.0040 0.7541 0.0127
FamEnv Fact ACC CA PDS Par 6658 13 Indirect -0.0011 -0.0109 -0.0060 0.0157 0.0025
FamEnv Fact ACC CA PDS Par 6658 13 Total 0.0144 -0.0344 -0.0100 0.4235 0.0125
FamEnv Fact ACC CA PDS Yth 6658 13 Apath -0.1214 -0.1719 -0.1467 0.0000 0.0129
FamEnv Fact ACC CA PDS Yth 6658 13 Bpath 0.0241 -0.0306 -0.0032 0.8162 0.0140
FamEnv Fact ACC CA PDS Yth 6658 13 Direct 0.0145 -0.0353 -0.0104 0.4125 0.0127
FamEnv Fact ACC CA PDS Yth 6658 13 Indirect 0.0045 -0.0035 0.0005 0.8163 0.0021
FamEnv Fact ACC CA PDS Yth 6658 13 Total 0.0146 -0.0345 -0.0099 0.4275 0.0125
FamEnv Fact ACC CT PDS ParYth 6658 13 Apath -0.1598 -0.2068 -0.1833 0.0000 0.0120
FamEnv Fact ACC CT PDS ParYth 6658 13 Bpath 0.0070 -0.0562 -0.0246 0.1276 0.0161
FamEnv Fact ACC CT PDS ParYth 6658 13 Direct 0.0100 -0.0416 -0.0158 0.2309 0.0132
FamEnv Fact ACC CT PDS ParYth 6658 13 Indirect 0.0103 -0.0013 0.0045 0.1279 0.0030
FamEnv Fact ACC CT PDS ParYth 6658 13 Total 0.0139 -0.0364 -0.0113 0.3803 0.0128
FamEnv Fact ACC CT PDS Par 6658 13 Apath -0.1423 -0.1829 -0.1626 0.0000 0.0104
FamEnv Fact ACC CT PDS Par 6658 13 Bpath -0.0188 -0.0787 -0.0487 0.0014 0.0153
FamEnv Fact ACC CT PDS Par 6658 13 Direct 0.0068 -0.0451 -0.0191 0.1481 0.0132
FamEnv Fact ACC CT PDS Par 6658 13 Indirect 0.0128 0.0030 0.0079 0.0016 0.0025
FamEnv Fact ACC CT PDS Par 6658 13 Total 0.0142 -0.0366 -0.0112 0.3877 0.0130
FamEnv Fact ACC CT PDS Yth 6658 13 Apath -0.1210 -0.1723 -0.1467 0.0000 0.0131
FamEnv Fact ACC CT PDS Yth 6658 13 Bpath 0.0272 -0.0271 0.0000 0.9987 0.0139
FamEnv Fact ACC CT PDS Yth 6658 13 Direct 0.0144 -0.0369 -0.0112 0.3900 0.0131
FamEnv Fact ACC CT PDS Yth 6658 13 Indirect 0.0040 -0.0040 0.0000 0.9987 0.0020
FamEnv Fact ACC CT PDS Yth 6658 13 Total 0.0140 -0.0365 -0.0112 0.3819 0.0129
FamEnv Fact Amygdala vol PDS ParYth 6658 13 Apath -0.1600 -0.2072 -0.1836 0.0000 0.0120
FamEnv Fact Amygdala vol PDS ParYth 6658 13 Bpath 0.0591 -0.0032 0.0280 0.0789 0.0159
FamEnv Fact Amygdala vol PDS ParYth 6658 13 Direct 0.0248 -0.0243 0.0002 0.9850 0.0125
FamEnv Fact Amygdala vol PDS ParYth 6658 13 Indirect 0.0007 -0.0109 -0.0051 0.0821 0.0030
FamEnv Fact Amygdala vol PDS ParYth 6658 13 Total 0.0189 -0.0287 -0.0049 0.6866 0.0121
FamEnv Fact Amygdala vol PDS Par 6658 13 Apath -0.1422 -0.1833 -0.1627 0.0000 0.0105
FamEnv Fact Amygdala vol PDS Par 6658 13 Bpath 0.0544 -0.0036 0.0254 0.0857 0.0148
FamEnv Fact Amygdala vol PDS Par 6658 13 Direct 0.0239 -0.0255 -0.0008 0.9497 0.0126
FamEnv Fact Amygdala vol PDS Par 6658 13 Indirect 0.0006 -0.0089 -0.0041 0.0872 0.0024
FamEnv Fact Amygdala vol PDS Par 6658 13 Total 0.0193 -0.0292 -0.0049 0.6898 0.0124
FamEnv Fact Amygdala vol PDS Yth 6658 13 Apath -0.1211 -0.1725 -0.1468 0.0000 0.0131
FamEnv Fact Amygdala vol PDS Yth 6658 13 Bpath 0.0377 -0.0189 0.0094 0.5153 0.0144
FamEnv Fact Amygdala vol PDS Yth 6658 13 Direct 0.0213 -0.0283 -0.0035 0.7803 0.0126
FamEnv Fact Amygdala vol PDS Yth 6658 13 Indirect 0.0028 -0.0055 -0.0014 0.5151 0.0021
FamEnv Fact Amygdala vol PDS Yth 6658 13 Total 0.0195 -0.0293 -0.0049 0.6931 0.0124
FamEnv Fact L AmygCON PDS ParYth 6658 13 Apath -0.1599 -0.2072 -0.1836 0.0000 0.0121
FamEnv Fact L AmygCON PDS ParYth 6658 13 Bpath -0.0131 -0.0783 -0.0457 0.0059 0.0166
FamEnv Fact L AmygCON PDS ParYth 6658 13 Direct 0.1243 0.0758 0.1000 0.0000 0.0124
FamEnv Fact L AmygCON PDS ParYth 6658 13 Indirect 0.0145 0.0023 0.0084 0.0070 0.0031
FamEnv Fact L AmygCON PDS ParYth 6658 13 Total 0.1320 0.0849 0.1084 0.0000 0.0120
FamEnv Fact L AmygCON PDS Par 6658 13 Apath -0.1424 -0.1836 -0.1630 0.0000 0.0105
FamEnv Fact L AmygCON PDS Par 6658 13 Bpath -0.0421 -0.1012 -0.0716 0.0000 0.0151
FamEnv Fact L AmygCON PDS Par 6658 13 Direct 0.1213 0.0728 0.0971 0.0000 0.0124
FamEnv Fact L AmygCON PDS Par 6658 13 Indirect 0.0168 0.0066 0.0117 0.0000 0.0026
FamEnv Fact L AmygCON PDS Par 6658 13 Total 0.1327 0.0849 0.1088 0.0000 0.0122
FamEnv Fact L AmygCON PDS Yth 6658 13 Apath -0.1211 -0.1722 -0.1467 0.0000 0.0130
FamEnv Fact L AmygCON PDS Yth 6658 13 Bpath 0.0163 -0.0431 -0.0134 0.3765 0.0151
FamEnv Fact L AmygCON PDS Yth 6658 13 Direct 0.1310 0.0818 0.1064 0.0000 0.0126
FamEnv Fact L AmygCON PDS Yth 6658 13 Indirect 0.0063 -0.0024 0.0020 0.3796 0.0022
FamEnv Fact L AmygCON PDS Yth 6658 13 Total 0.1323 0.0844 0.1084 0.0000 0.0122
FamEnv Fact R AmygCON PDS ParYth 6658 13 Apath -0.1602 -0.2067 -0.1834 0.0000 0.0118
FamEnv Fact R AmygCON PDS ParYth 6658 13 Bpath -0.0107 -0.0735 -0.0421 0.0085 0.0160
FamEnv Fact R AmygCON PDS ParYth 6658 13 Direct 0.1027 0.0520 0.0774 0.0000 0.0129
FamEnv Fact R AmygCON PDS ParYth 6658 13 Indirect 0.0136 0.0019 0.0077 0.0098 0.0030
FamEnv Fact R AmygCON PDS ParYth 6658 13 Total 0.1097 0.0605 0.0851 0.0000 0.0126
FamEnv Fact R AmygCON PDS Par 6658 13 Apath -0.1423 -0.1834 -0.1628 0.0000 0.0105
FamEnv Fact R AmygCON PDS Par 6658 13 Bpath -0.0358 -0.0940 -0.0649 0.0000 0.0148
FamEnv Fact R AmygCON PDS Par 6658 13 Direct 0.0996 0.0501 0.0748 0.0000 0.0126
FamEnv Fact R AmygCON PDS Par 6658 13 Indirect 0.0155 0.0057 0.0106 0.0000 0.0025
FamEnv Fact R AmygCON PDS Par 6658 13 Total 0.1099 0.0609 0.0854 0.0000 0.0125
FamEnv Fact R AmygCON PDS Yth 6658 13 Apath -0.1210 -0.1723 -0.1466 0.0000 0.0131
FamEnv Fact R AmygCON PDS Yth 6658 13 Bpath 0.0161 -0.0414 -0.0127 0.3878 0.0147
FamEnv Fact R AmygCON PDS Yth 6658 13 Direct 0.1084 0.0580 0.0832 0.0000 0.0129
FamEnv Fact R AmygCON PDS Yth 6658 13 Indirect 0.0061 -0.0024 0.0019 0.3900 0.0022
FamEnv Fact R AmygCON PDS Yth 6658 13 Total 0.1097 0.0604 0.0850 0.0000 0.0126
FES Par ACC CA PDS ParYth 6653 13 Apath -0.0068 -0.0546 -0.0307 0.0118 0.0122
FES Par ACC CA PDS ParYth 6653 13 Bpath 0.0176 -0.0424 -0.0124 0.4167 0.0153
FES Par ACC CA PDS ParYth 6653 13 Direct 0.0231 -0.0265 -0.0017 0.8917 0.0126
FES Par ACC CA PDS ParYth 6653 13 Indirect 0.0006 -0.0013 -0.0004 0.4362 0.0005
FES Par ACC CA PDS ParYth 6653 13 Total 0.0227 -0.0269 -0.0021 0.8679 0.0126
FES Par ACC CA PDS Par 6653 13 Apath -0.0223 -0.0637 -0.0430 0.0000 0.0106
FES Par ACC CA PDS Par 6653 13 Bpath -0.0096 -0.0667 -0.0382 0.0088 0.0146
FES Par ACC CA PDS Par 6653 13 Direct 0.0244 -0.0254 -0.0005 0.9675 0.0127
FES Par ACC CA PDS Par 6653 13 Indirect -0.0002 -0.0031 -0.0016 0.0287 0.0008
FES Par ACC CA PDS Par 6653 13 Total 0.0228 -0.0271 -0.0022 0.8652 0.0127
FES Par ACC CA PDS Yth 6653 13 Apath 0.0208 -0.0315 -0.0054 0.6867 0.0133
FES Par ACC CA PDS Yth 6653 13 Bpath 0.0287 -0.0253 0.0017 0.9013 0.0138
FES Par ACC CA PDS Yth 6653 13 Direct 0.0223 -0.0265 -0.0021 0.8652 0.0125
FES Par ACC CA PDS Yth 6653 13 Indirect 0.0000 0.0000 0.0000 0.0000 0.0000
FES Par ACC CA PDS Yth 6653 13 Total 0.0223 -0.0265 -0.0021 0.8658 0.0125
FES Par ACC CT PDS ParYth 6653 13 Apath -0.0064 -0.0548 -0.0306 0.0133 0.0124
FES Par ACC CT PDS ParYth 6653 13 Bpath 0.0538 -0.0087 0.0226 0.1566 0.0159
FES Par ACC CT PDS ParYth 6653 13 Direct 0.0240 -0.0251 -0.0005 0.9654 0.0125
FES Par ACC CT PDS ParYth 6653 13 Indirect 0.0018 -0.0004 0.0007 0.2135 0.0006
FES Par ACC CT PDS ParYth 6653 13 Total 0.0247 -0.0244 0.0001 0.9906 0.0125
FES Par ACC CT PDS Par 6653 13 Apath -0.0224 -0.0638 -0.0431 0.0000 0.0106
FES Par ACC CT PDS Par 6653 13 Bpath 0.0747 0.0167 0.0457 0.0020 0.0148
FES Par ACC CT PDS Par 6653 13 Direct 0.0228 -0.0264 -0.0018 0.8889 0.0126
FES Par ACC CT PDS Par 6653 13 Indirect 0.0035 0.0004 0.0020 0.0131 0.0008
FES Par ACC CT PDS Par 6653 13 Total 0.0248 -0.0244 0.0002 0.9861 0.0125
FES Par ACC CT PDS Yth 6653 13 Apath 0.0213 -0.0321 -0.0054 0.6935 0.0136
FES Par ACC CT PDS Yth 6653 13 Bpath 0.0265 -0.0279 -0.0007 0.9593 0.0139
FES Par ACC CT PDS Yth 6653 13 Direct 0.0242 -0.0239 0.0002 0.9898 0.0123
FES Par ACC CT PDS Yth 6653 13 Indirect 0.0000 0.0000 0.0000 0.0000 0.0000
FES Par ACC CT PDS Yth 6653 13 Total 0.0242 -0.0239 0.0002 0.9900 0.0123
FES Par Amygdala vol PDS ParYth 6653 13 Apath -0.0067 -0.0551 -0.0309 0.0122 0.0123
FES Par Amygdala vol PDS ParYth 6653 13 Bpath 0.0055 -0.0552 -0.0248 0.1084 0.0155
FES Par Amygdala vol PDS ParYth 6653 13 Direct 0.0080 -0.0386 -0.0153 0.1973 0.0119
FES Par Amygdala vol PDS ParYth 6653 13 Indirect 0.0003 -0.0019 -0.0008 0.1763 0.0006
FES Par Amygdala vol PDS ParYth 6653 13 Total 0.0072 -0.0394 -0.0161 0.1753 0.0119
FES Par Amygdala vol PDS Par 6653 13 Apath -0.0227 -0.0636 -0.0431 0.0000 0.0104
FES Par Amygdala vol PDS Par 6653 13 Bpath 0.0037 -0.0531 -0.0247 0.0881 0.0145
FES Par Amygdala vol PDS Par 6653 13 Direct 0.0082 -0.0384 -0.0151 0.2045 0.0119
FES Par Amygdala vol PDS Par 6653 13 Indirect 0.0003 -0.0024 -0.0011 0.1144 0.0007
FES Par Amygdala vol PDS Par 6653 13 Total 0.0071 -0.0394 -0.0161 0.1740 0.0119
FES Par Amygdala vol PDS Yth 6653 13 Apath 0.0209 -0.0317 -0.0054 0.6886 0.0134
FES Par Amygdala vol PDS Yth 6653 13 Bpath 0.0199 -0.0347 -0.0074 0.5964 0.0139
FES Par Amygdala vol PDS Yth 6653 13 Direct 0.0075 -0.0396 -0.0161 0.1812 0.0120
FES Par Amygdala vol PDS Yth 6653 13 Indirect 0.0002 -0.0003 0.0000 0.7481 0.0001
FES Par Amygdala vol PDS Yth 6653 13 Total 0.0074 -0.0396 -0.0161 0.1801 0.0120
FES Par L AmygCON PDS ParYth 6653 13 Apath -0.0063 -0.0542 -0.0303 0.0132 0.0122
FES Par L AmygCON PDS ParYth 6653 13 Bpath 0.0974 0.0351 0.0663 0.0000 0.0159
FES Par L AmygCON PDS ParYth 6653 13 Direct 0.0424 -0.0051 0.0187 0.1229 0.0121
FES Par L AmygCON PDS ParYth 6653 13 Indirect 0.0038 0.0002 0.0020 0.0285 0.0009
FES Par L AmygCON PDS ParYth 6653 13 Total 0.0445 -0.0031 0.0207 0.0878 0.0121
FES Par L AmygCON PDS Par 6653 13 Apath -0.0226 -0.0640 -0.0433 0.0000 0.0106
FES Par L AmygCON PDS Par 6653 13 Bpath 0.1194 0.0605 0.0899 0.0000 0.0150
FES Par L AmygCON PDS Par 6653 13 Direct 0.0408 -0.0064 0.0172 0.1532 0.0120
FES Par L AmygCON PDS Par 6653 13 Indirect 0.0061 0.0017 0.0039 0.0006 0.0011
FES Par L AmygCON PDS Par 6653 13 Total 0.0447 -0.0025 0.0211 0.0801 0.0120
FES Par L AmygCON PDS Yth 6653 13 Apath 0.0209 -0.0310 -0.0051 0.7019 0.0132
FES Par L AmygCON PDS Yth 6653 13 Bpath 0.0581 0.0005 0.0293 0.0465 0.0147
FES Par L AmygCON PDS Yth 6653 13 Direct 0.0441 -0.0030 0.0205 0.0877 0.0120
FES Par L AmygCON PDS Yth 6653 13 Indirect 0.0009 -0.0006 0.0001 0.7056 0.0004
FES Par L AmygCON PDS Yth 6653 13 Total 0.0443 -0.0029 0.0207 0.0857 0.0120
FES Par R AmygCON PDS ParYth 6653 13 Apath -0.0061 -0.0544 -0.0303 0.0139 0.0123
FES Par R AmygCON PDS ParYth 6653 13 Bpath 0.0902 0.0281 0.0592 0.0002 0.0158
FES Par R AmygCON PDS ParYth 6653 13 Direct 0.0299 -0.0187 0.0056 0.6495 0.0124
FES Par R AmygCON PDS ParYth 6653 13 Indirect 0.0035 0.0001 0.0018 0.0400 0.0009
FES Par R AmygCON PDS ParYth 6653 13 Total 0.0318 -0.0169 0.0074 0.5498 0.0124
FES Par R AmygCON PDS Par 6653 13 Apath -0.0221 -0.0639 -0.0430 0.0001 0.0107
FES Par R AmygCON PDS Par 6653 13 Bpath 0.1084 0.0512 0.0798 0.0000 0.0146
FES Par R AmygCON PDS Par 6653 13 Direct 0.0285 -0.0198 0.0043 0.7247 0.0123
FES Par R AmygCON PDS Par 6653 13 Indirect 0.0055 0.0014 0.0034 0.0010 0.0010
FES Par R AmygCON PDS Par 6653 13 Total 0.0319 -0.0164 0.0078 0.5290 0.0123
FES Par R AmygCON PDS Yth 6653 13 Apath 0.0208 -0.0311 -0.0052 0.6967 0.0132
FES Par R AmygCON PDS Yth 6653 13 Bpath 0.0539 -0.0033 0.0253 0.0830 0.0146
FES Par R AmygCON PDS Yth 6653 13 Direct 0.0313 -0.0167 0.0073 0.5518 0.0123
FES Par R AmygCON PDS Yth 6653 13 Indirect 0.0008 -0.0005 0.0001 0.7035 0.0003
FES Par R AmygCON PDS Yth 6653 13 Total 0.0315 -0.0166 0.0074 0.5448 0.0123
FES Yth ACC CA PDS ParYth 6640 13 Apath -0.0534 -0.1033 -0.0784 0.0000 0.0127
FES Yth ACC CA PDS ParYth 6640 13 Bpath 0.0165 -0.0429 -0.0132 0.3840 0.0151
FES Yth ACC CA PDS ParYth 6640 13 Direct 0.0190 -0.0306 -0.0058 0.6454 0.0126
FES Yth ACC CA PDS ParYth 6640 13 Indirect 0.0013 -0.0034 -0.0010 0.3883 0.0012
FES Yth ACC CA PDS ParYth 6640 13 Total 0.0178 -0.0315 -0.0068 0.5865 0.0126
FES Yth ACC CA PDS Par 6640 13 Apath -0.0175 -0.0590 -0.0382 0.0003 0.0106
FES Yth ACC CA PDS Par 6640 13 Bpath -0.0088 -0.0668 -0.0378 0.0107 0.0148
FES Yth ACC CA PDS Par 6640 13 Direct 0.0186 -0.0295 -0.0054 0.6573 0.0123
FES Yth ACC CA PDS Par 6640 13 Indirect -0.0001 -0.0028 -0.0014 0.0345 0.0007
FES Yth ACC CA PDS Par 6640 13 Total 0.0171 -0.0309 -0.0069 0.5743 0.0123
FES Yth ACC CA PDS Yth 6640 13 Apath -0.0720 -0.1252 -0.0986 0.0000 0.0136
FES Yth ACC CA PDS Yth 6640 13 Bpath 0.0288 -0.0255 0.0016 0.9069 0.0139
FES Yth ACC CA PDS Yth 6640 13 Direct 0.0174 -0.0314 -0.0070 0.5743 0.0125
FES Yth ACC CA PDS Yth 6640 13 Indirect 0.0028 -0.0025 0.0002 0.9070 0.0014
FES Yth ACC CA PDS Yth 6640 13 Total 0.0175 -0.0311 -0.0068 0.5811 0.0124
FES Yth ACC CT PDS ParYth 6640 13 Apath -0.0539 -0.1028 -0.0783 0.0000 0.0125
FES Yth ACC CT PDS ParYth 6640 13 Bpath 0.0544 -0.0079 0.0233 0.1431 0.0159
FES Yth ACC CT PDS ParYth 6640 13 Direct 0.0179 -0.0321 -0.0071 0.5759 0.0128
FES Yth ACC CT PDS ParYth 6640 13 Indirect 0.0043 -0.0007 0.0018 0.1508 0.0013
FES Yth ACC CT PDS ParYth 6640 13 Total 0.0196 -0.0302 -0.0053 0.6762 0.0127
FES Yth ACC CT PDS Par 6640 13 Apath -0.0168 -0.0593 -0.0381 0.0004 0.0108
FES Yth ACC CT PDS Par 6640 13 Bpath 0.0744 0.0159 0.0452 0.0025 0.0149
FES Yth ACC CT PDS Par 6640 13 Direct 0.0177 -0.0316 -0.0070 0.5785 0.0126
FES Yth ACC CT PDS Par 6640 13 Indirect 0.0032 0.0003 0.0017 0.0203 0.0007
FES Yth ACC CT PDS Par 6640 13 Total 0.0194 -0.0299 -0.0053 0.6754 0.0126
FES Yth ACC CT PDS Yth 6640 13 Apath -0.0722 -0.1251 -0.0986 0.0000 0.0135
FES Yth ACC CT PDS Yth 6640 13 Bpath 0.0277 -0.0280 -0.0001 0.9924 0.0142
FES Yth ACC CT PDS Yth 6640 13 Direct 0.0195 -0.0302 -0.0053 0.6754 0.0127
FES Yth ACC CT PDS Yth 6640 13 Indirect 0.0027 -0.0028 0.0000 0.9924 0.0014
FES Yth ACC CT PDS Yth 6640 13 Total 0.0194 -0.0301 -0.0053 0.6733 0.0126
FES Yth Amygdala vol PDS ParYth 6640 13 Apath -0.0539 -0.1027 -0.0783 0.0000 0.0124
FES Yth Amygdala vol PDS ParYth 6640 13 Bpath 0.0050 -0.0568 -0.0259 0.0998 0.0158
FES Yth Amygdala vol PDS ParYth 6640 13 Direct 0.0272 -0.0203 0.0034 0.7762 0.0121
FES Yth Amygdala vol PDS ParYth 6640 13 Indirect 0.0005 -0.0045 -0.0020 0.1116 0.0013
FES Yth Amygdala vol PDS ParYth 6640 13 Total 0.0249 -0.0221 0.0014 0.9063 0.0120
FES Yth Amygdala vol PDS Par 6640 13 Apath -0.0174 -0.0590 -0.0382 0.0003 0.0106
FES Yth Amygdala vol PDS Par 6640 13 Bpath 0.0039 -0.0530 -0.0245 0.0904 0.0145
FES Yth Amygdala vol PDS Par 6640 13 Direct 0.0261 -0.0214 0.0023 0.8472 0.0121
FES Yth Amygdala vol PDS Par 6640 13 Indirect 0.0003 -0.0021 -0.0009 0.1222 0.0006
FES Yth Amygdala vol PDS Par 6640 13 Total 0.0251 -0.0223 0.0014 0.9082 0.0121
FES Yth Amygdala vol PDS Yth 6640 13 Apath -0.0724 -0.1249 -0.0987 0.0000 0.0134
FES Yth Amygdala vol PDS Yth 6640 13 Bpath 0.0191 -0.0368 -0.0089 0.5342 0.0143
FES Yth Amygdala vol PDS Yth 6640 13 Direct 0.0262 -0.0216 0.0023 0.8506 0.0122
FES Yth Amygdala vol PDS Yth 6640 13 Indirect 0.0019 -0.0036 -0.0009 0.5346 0.0014
FES Yth Amygdala vol PDS Yth 6640 13 Total 0.0251 -0.0223 0.0014 0.9063 0.0121
FES Yth L AmygCON PDS ParYth 6640 13 Apath -0.0539 -0.1033 -0.0786 0.0000 0.0126
FES Yth L AmygCON PDS ParYth 6640 13 Bpath 0.0971 0.0330 0.0651 0.0001 0.0163
FES Yth L AmygCON PDS ParYth 6640 13 Direct 0.0637 0.0147 0.0392 0.0017 0.0125
FES Yth L AmygCON PDS ParYth 6640 13 Indirect 0.0081 0.0021 0.0051 0.0007 0.0015
FES Yth L AmygCON PDS ParYth 6640 13 Total 0.0686 0.0200 0.0443 0.0003 0.0124
FES Yth L AmygCON PDS Par 6640 13 Apath -0.0174 -0.0591 -0.0383 0.0003 0.0106
FES Yth L AmygCON PDS Par 6640 13 Bpath 0.1193 0.0594 0.0893 0.0000 0.0153
FES Yth L AmygCON PDS Par 6640 13 Direct 0.0657 0.0164 0.0411 0.0011 0.0126
FES Yth L AmygCON PDS Par 6640 13 Indirect 0.0056 0.0012 0.0034 0.0020 0.0011
FES Yth L AmygCON PDS Par 6640 13 Total 0.0692 0.0197 0.0445 0.0004 0.0126
FES Yth L AmygCON PDS Yth 6640 13 Apath -0.0723 -0.1253 -0.0988 0.0000 0.0135
FES Yth L AmygCON PDS Yth 6640 13 Bpath 0.0550 -0.0031 0.0260 0.0795 0.0148
FES Yth L AmygCON PDS Yth 6640 13 Direct 0.0672 0.0158 0.0415 0.0016 0.0131
FES Yth L AmygCON PDS Yth 6640 13 Indirect 0.0055 -0.0004 0.0026 0.0878 0.0015
FES Yth L AmygCON PDS Yth 6640 13 Total 0.0695 0.0187 0.0441 0.0007 0.0130
FES Yth R AmygCON PDS ParYth 6640 13 Apath -0.0542 -0.1024 -0.0783 0.0000 0.0123
FES Yth R AmygCON PDS ParYth 6640 13 Bpath 0.0874 0.0253 0.0563 0.0004 0.0159
FES Yth R AmygCON PDS ParYth 6640 13 Direct 0.0760 0.0222 0.0491 0.0003 0.0137
FES Yth R AmygCON PDS ParYth 6640 13 Indirect 0.0072 0.0016 0.0044 0.0018 0.0014
FES Yth R AmygCON PDS ParYth 6640 13 Total 0.0803 0.0267 0.0535 0.0001 0.0137
FES Yth R AmygCON PDS Par 6640 13 Apath -0.0173 -0.0595 -0.0384 0.0004 0.0108
FES Yth R AmygCON PDS Par 6640 13 Bpath 0.1072 0.0487 0.0779 0.0000 0.0149
FES Yth R AmygCON PDS Par 6640 13 Direct 0.0778 0.0235 0.0507 0.0003 0.0138
FES Yth R AmygCON PDS Par 6640 13 Indirect 0.0050 0.0010 0.0030 0.0034 0.0010
FES Yth R AmygCON PDS Par 6640 13 Total 0.0809 0.0265 0.0537 0.0001 0.0139
FES Yth R AmygCON PDS Yth 6640 13 Apath -0.0722 -0.1252 -0.0987 0.0000 0.0135
FES Yth R AmygCON PDS Yth 6640 13 Bpath 0.0510 -0.0089 0.0211 0.1683 0.0153
FES Yth R AmygCON PDS Yth 6640 13 Direct 0.0783 0.0242 0.0512 0.0002 0.0138
FES Yth R AmygCON PDS Yth 6640 13 Indirect 0.0051 -0.0009 0.0021 0.1762 0.0015
FES Yth R AmygCON PDS Yth 6640 13 Total 0.0800 0.0266 0.0533 0.0001 0.0136
Par Fact ACC CA PDS ParYth 6656 13 Apath -0.0239 -0.0723 -0.0481 0.0001 0.0123
Par Fact ACC CA PDS ParYth 6656 13 Bpath 0.0411 -0.0177 0.0117 0.4351 0.0150
Par Fact ACC CA PDS ParYth 6656 13 Direct 0.0219 -0.0270 -0.0025 0.8389 0.0125
Par Fact ACC CA PDS ParYth 6656 13 Indirect 0.0009 -0.0020 -0.0006 0.4430 0.0007
Par Fact ACC CA PDS ParYth 6656 13 Total 0.0213 -0.0275 -0.0031 0.8034 0.0125
Par Fact ACC CA PDS Par 6656 13 Apath -0.0359 -0.0767 -0.0563 0.0000 0.0104
Par Fact ACC CA PDS Par 6656 13 Bpath 0.0664 0.0090 0.0377 0.0101 0.0146
Par Fact ACC CA PDS Par 6656 13 Direct 0.0237 -0.0258 -0.0010 0.9351 0.0126
Par Fact ACC CA PDS Par 6656 13 Indirect -0.0003 -0.0039 -0.0021 0.0204 0.0009
Par Fact ACC CA PDS Par 6656 13 Total 0.0215 -0.0278 -0.0031 0.8026 0.0126
Par Fact ACC CA PDS Yth 6656 13 Apath 0.0044 -0.0481 -0.0218 0.1031 0.0134
Par Fact ACC CA PDS Yth 6656 13 Bpath 0.0251 -0.0294 -0.0021 0.8776 0.0139
Par Fact ACC CA PDS Yth 6656 13 Direct 0.0213 -0.0276 -0.0031 0.8011 0.0125
Par Fact ACC CA PDS Yth 6656 13 Indirect 0.0006 -0.0006 0.0000 0.8787 0.0003
Par Fact ACC CA PDS Yth 6656 13 Total 0.0214 -0.0276 -0.0031 0.8040 0.0125
Par Fact ACC CT PDS ParYth 6656 13 Apath -0.0238 -0.0720 -0.0479 0.0001 0.0123
Par Fact ACC CT PDS ParYth 6656 13 Bpath 0.0085 -0.0530 -0.0222 0.1559 0.0157
Par Fact ACC CT PDS ParYth 6656 13 Direct 0.0220 -0.0274 -0.0027 0.8321 0.0126
Par Fact ACC CT PDS ParYth 6656 13 Indirect 0.0026 -0.0005 0.0011 0.1775 0.0008
Par Fact ACC CT PDS ParYth 6656 13 Total 0.0231 -0.0263 -0.0016 0.8985 0.0126
Par Fact ACC CT PDS Par 6656 13 Apath -0.0360 -0.0766 -0.0563 0.0000 0.0104
Par Fact ACC CT PDS Par 6656 13 Bpath -0.0173 -0.0745 -0.0459 0.0017 0.0146
Par Fact ACC CT PDS Par 6656 13 Direct 0.0200 -0.0283 -0.0041 0.7374 0.0123
Par Fact ACC CT PDS Par 6656 13 Indirect 0.0044 0.0007 0.0026 0.0063 0.0009
Par Fact ACC CT PDS Par 6656 13 Total 0.0226 -0.0257 -0.0015 0.8998 0.0123
Par Fact ACC CT PDS Yth 6656 13 Apath 0.0036 -0.0473 -0.0218 0.0927 0.0130
Par Fact ACC CT PDS Yth 6656 13 Bpath 0.0282 -0.0261 0.0011 0.9392 0.0138
Par Fact ACC CT PDS Yth 6656 13 Direct 0.0227 -0.0258 -0.0016 0.8983 0.0124
Par Fact ACC CT PDS Yth 6656 13 Indirect 0.0006 -0.0006 0.0000 0.9394 0.0003
Par Fact ACC CT PDS Yth 6656 13 Total 0.0226 -0.0258 -0.0016 0.8967 0.0124
Par Fact Amygdala vol PDS ParYth 6656 13 Apath -0.0248 -0.0718 -0.0483 0.0001 0.0120
Par Fact Amygdala vol PDS ParYth 6656 13 Bpath 0.0566 -0.0055 0.0255 0.1069 0.0158
Par Fact Amygdala vol PDS ParYth 6656 13 Direct 0.0092 -0.0379 -0.0144 0.2308 0.0120
Par Fact Amygdala vol PDS ParYth 6656 13 Indirect 0.0004 -0.0028 -0.0012 0.1349 0.0008
Par Fact Amygdala vol PDS ParYth 6656 13 Total 0.0079 -0.0391 -0.0156 0.1924 0.0120
Par Fact Amygdala vol PDS Par 6656 13 Apath -0.0359 -0.0768 -0.0564 0.0000 0.0104
Par Fact Amygdala vol PDS Par 6656 13 Bpath 0.0534 -0.0037 0.0248 0.0883 0.0146
Par Fact Amygdala vol PDS Par 6656 13 Direct 0.0092 -0.0378 -0.0143 0.2344 0.0120
Par Fact Amygdala vol PDS Par 6656 13 Indirect 0.0003 -0.0031 -0.0014 0.1021 0.0009
Par Fact Amygdala vol PDS Par 6656 13 Total 0.0078 -0.0391 -0.0157 0.1908 0.0120
Par Fact Amygdala vol PDS Yth 6656 13 Apath 0.0040 -0.0476 -0.0218 0.0970 0.0132
Par Fact Amygdala vol PDS Yth 6656 13 Bpath 0.0352 -0.0191 0.0080 0.5621 0.0139
Par Fact Amygdala vol PDS Yth 6656 13 Direct 0.0078 -0.0387 -0.0155 0.1935 0.0119
Par Fact Amygdala vol PDS Yth 6656 13 Indirect 0.0004 -0.0008 -0.0002 0.5806 0.0003
Par Fact Amygdala vol PDS Yth 6656 13 Total 0.0077 -0.0389 -0.0156 0.1885 0.0119
Par Fact L AmygCON PDS ParYth 6656 13 Apath -0.0241 -0.0717 -0.0479 0.0001 0.0121
Par Fact L AmygCON PDS ParYth 6656 13 Bpath -0.0343 -0.0974 -0.0659 0.0000 0.0161
Par Fact L AmygCON PDS ParYth 6656 13 Direct 0.0480 0.0003 0.0241 0.0472 0.0122
Par Fact L AmygCON PDS ParYth 6656 13 Indirect 0.0053 0.0010 0.0032 0.0036 0.0011
Par Fact L AmygCON PDS ParYth 6656 13 Total 0.0511 0.0034 0.0273 0.0249 0.0122
Par Fact L AmygCON PDS Par 6656 13 Apath -0.0360 -0.0772 -0.0566 0.0000 0.0105
Par Fact L AmygCON PDS Par 6656 13 Bpath -0.0593 -0.1195 -0.0894 0.0000 0.0154
Par Fact L AmygCON PDS Par 6656 13 Direct 0.0463 -0.0012 0.0225 0.0632 0.0121
Par Fact L AmygCON PDS Par 6656 13 Indirect 0.0076 0.0025 0.0051 0.0001 0.0013
Par Fact L AmygCON PDS Par 6656 13 Total 0.0514 0.0038 0.0276 0.0232 0.0122
Par Fact L AmygCON PDS Yth 6656 13 Apath 0.0040 -0.0473 -0.0216 0.0986 0.0131
Par Fact L AmygCON PDS Yth 6656 13 Bpath -0.0007 -0.0574 -0.0290 0.0450 0.0145
Par Fact L AmygCON PDS Yth 6656 13 Direct 0.0503 0.0031 0.0267 0.0266 0.0120
Par Fact L AmygCON PDS Yth 6656 13 Indirect 0.0016 -0.0003 0.0006 0.2002 0.0005
Par Fact L AmygCON PDS Yth 6656 13 Total 0.0509 0.0037 0.0273 0.0232 0.0120
Par Fact R AmygCON PDS ParYth 6656 13 Apath -0.0237 -0.0717 -0.0477 0.0001 0.0122
Par Fact R AmygCON PDS ParYth 6656 13 Bpath -0.0282 -0.0891 -0.0586 0.0002 0.0155
Par Fact R AmygCON PDS ParYth 6656 13 Direct 0.0334 -0.0136 0.0099 0.4100 0.0120
Par Fact R AmygCON PDS ParYth 6656 13 Indirect 0.0048 0.0008 0.0028 0.0052 0.0010
Par Fact R AmygCON PDS ParYth 6656 13 Total 0.0362 -0.0108 0.0127 0.2903 0.0120
Par Fact R AmygCON PDS Par 6656 13 Apath -0.0355 -0.0770 -0.0563 0.0000 0.0106
Par Fact R AmygCON PDS Par 6656 13 Bpath -0.0509 -0.1080 -0.0794 0.0000 0.0146
Par Fact R AmygCON PDS Par 6656 13 Direct 0.0321 -0.0151 0.0085 0.4813 0.0120
Par Fact R AmygCON PDS Par 6656 13 Indirect 0.0068 0.0022 0.0045 0.0001 0.0012
Par Fact R AmygCON PDS Par 6656 13 Total 0.0366 -0.0107 0.0129 0.2828 0.0121
Par Fact R AmygCON PDS Yth 6656 13 Apath 0.0043 -0.0475 -0.0216 0.1021 0.0132
Par Fact R AmygCON PDS Yth 6656 13 Bpath 0.0037 -0.0537 -0.0250 0.0872 0.0146
Par Fact R AmygCON PDS Yth 6656 13 Direct 0.0362 -0.0119 0.0122 0.3211 0.0123
Par Fact R AmygCON PDS Yth 6656 13 Indirect 0.0015 -0.0004 0.0005 0.2436 0.0005
Par Fact R AmygCON PDS Yth 6656 13 Total 0.0367 -0.0113 0.0127 0.2995 0.0123
PMON ACC CA PDS ParYth 6644 13 Apath -0.0524 -0.1028 -0.0776 0.0000 0.0128
PMON ACC CA PDS ParYth 6644 13 Bpath 0.0437 -0.0160 0.0139 0.3634 0.0152
PMON ACC CA PDS ParYth 6644 13 Direct 0.0482 -0.0005 0.0238 0.0552 0.0124
PMON ACC CA PDS ParYth 6644 13 Indirect 0.0013 -0.0034 -0.0011 0.3681 0.0012
PMON ACC CA PDS ParYth 6644 13 Total 0.0470 -0.0015 0.0228 0.0659 0.0124
PMON ACC CA PDS Par 6644 13 Apath -0.0251 -0.0684 -0.0468 0.0000 0.0110
PMON ACC CA PDS Par 6644 13 Bpath 0.0681 0.0108 0.0395 0.0069 0.0146
PMON ACC CA PDS Par 6644 13 Direct 0.0490 0.0001 0.0246 0.0489 0.0125
PMON ACC CA PDS Par 6644 13 Indirect -0.0003 -0.0034 -0.0018 0.0232 0.0008
PMON ACC CA PDS Par 6644 13 Total 0.0471 -0.0017 0.0227 0.0679 0.0124
PMON ACC CA PDS Yth 6644 13 Apath -0.0502 -0.1050 -0.0776 0.0000 0.0140
PMON ACC CA PDS Yth 6644 13 Bpath 0.0265 -0.0288 -0.0012 0.9345 0.0141
PMON ACC CA PDS Yth 6644 13 Direct 0.0475 -0.0022 0.0226 0.0742 0.0127
PMON ACC CA PDS Yth 6644 13 Indirect 0.0022 -0.0021 0.0001 0.9345 0.0011
PMON ACC CA PDS Yth 6644 13 Total 0.0475 -0.0020 0.0227 0.0716 0.0126
PMON ACC CT PDS ParYth 6644 13 Apath -0.0523 -0.1029 -0.0776 0.0000 0.0129
PMON ACC CT PDS ParYth 6644 13 Bpath 0.0051 -0.0575 -0.0262 0.1005 0.0159
PMON ACC CT PDS ParYth 6644 13 Direct -0.0117 -0.0624 -0.0371 0.0041 0.0129
PMON ACC CT PDS ParYth 6644 13 Indirect 0.0045 -0.0005 0.0020 0.1112 0.0013
PMON ACC CT PDS ParYth 6644 13 Total -0.0098 -0.0603 -0.0350 0.0065 0.0129
PMON ACC CT PDS Par 6644 13 Apath -0.0252 -0.0685 -0.0468 0.0000 0.0111
PMON ACC CT PDS Par 6644 13 Bpath -0.0194 -0.0764 -0.0479 0.0010 0.0145
PMON ACC CT PDS Par 6644 13 Direct -0.0124 -0.0620 -0.0372 0.0032 0.0126
PMON ACC CT PDS Par 6644 13 Indirect 0.0039 0.0006 0.0022 0.0078 0.0008
PMON ACC CT PDS Par 6644 13 Total -0.0102 -0.0598 -0.0350 0.0057 0.0126
PMON ACC CT PDS Yth 6644 13 Apath -0.0509 -0.1043 -0.0776 0.0000 0.0136
PMON ACC CT PDS Yth 6644 13 Bpath 0.0257 -0.0282 -0.0013 0.9268 0.0137
PMON ACC CT PDS Yth 6644 13 Direct -0.0101 -0.0601 -0.0351 0.0059 0.0128
PMON ACC CT PDS Yth 6644 13 Indirect 0.0022 -0.0020 0.0001 0.9268 0.0011
PMON ACC CT PDS Yth 6644 13 Total -0.0100 -0.0599 -0.0350 0.0060 0.0127
PMON Amygdala vol PDS ParYth 6644 13 Apath -0.0530 -0.1022 -0.0776 0.0000 0.0126
PMON Amygdala vol PDS ParYth 6644 13 Bpath 0.0571 -0.0041 0.0265 0.0893 0.0156
PMON Amygdala vol PDS ParYth 6644 13 Direct 0.0238 -0.0251 -0.0006 0.9613 0.0125
PMON Amygdala vol PDS ParYth 6644 13 Indirect 0.0004 -0.0046 -0.0021 0.1065 0.0013
PMON Amygdala vol PDS ParYth 6644 13 Total 0.0216 -0.0269 -0.0027 0.8297 0.0124
PMON Amygdala vol PDS Par 6644 13 Apath -0.0248 -0.0687 -0.0467 0.0000 0.0112
PMON Amygdala vol PDS Par 6644 13 Bpath 0.0529 -0.0029 0.0250 0.0786 0.0142
PMON Amygdala vol PDS Par 6644 13 Direct 0.0227 -0.0258 -0.0015 0.9005 0.0124
PMON Amygdala vol PDS Par 6644 13 Indirect 0.0003 -0.0026 -0.0012 0.1068 0.0007
PMON Amygdala vol PDS Par 6644 13 Total 0.0214 -0.0269 -0.0027 0.8255 0.0123
PMON Amygdala vol PDS Yth 6644 13 Apath -0.0504 -0.1049 -0.0776 0.0000 0.0139
PMON Amygdala vol PDS Yth 6644 13 Bpath 0.0368 -0.0178 0.0095 0.4967 0.0139
PMON Amygdala vol PDS Yth 6644 13 Direct 0.0224 -0.0264 -0.0020 0.8749 0.0124
PMON Amygdala vol PDS Yth 6644 13 Indirect 0.0014 -0.0029 -0.0007 0.5006 0.0011
PMON Amygdala vol PDS Yth 6644 13 Total 0.0216 -0.0270 -0.0027 0.8282 0.0124
PMON L AmygCON PDS ParYth 6644 13 Apath -0.0523 -0.1031 -0.0777 0.0000 0.0130
PMON L AmygCON PDS ParYth 6644 13 Bpath -0.0329 -0.0961 -0.0645 0.0001 0.0161
PMON L AmygCON PDS ParYth 6644 13 Direct 0.0599 0.0079 0.0339 0.0107 0.0133
PMON L AmygCON PDS ParYth 6644 13 Indirect 0.0080 0.0021 0.0050 0.0008 0.0015
PMON L AmygCON PDS ParYth 6644 13 Total 0.0646 0.0131 0.0389 0.0031 0.0131
PMON L AmygCON PDS Par 6644 13 Apath -0.0248 -0.0689 -0.0469 0.0000 0.0113
PMON L AmygCON PDS Par 6644 13 Bpath -0.0586 -0.1192 -0.0889 0.0000 0.0154
PMON L AmygCON PDS Par 6644 13 Direct 0.0611 0.0092 0.0352 0.0079 0.0132
PMON L AmygCON PDS Par 6644 13 Indirect 0.0066 0.0018 0.0042 0.0007 0.0012
PMON L AmygCON PDS Par 6644 13 Total 0.0653 0.0134 0.0393 0.0030 0.0132
PMON L AmygCON PDS Yth 6644 13 Apath -0.0505 -0.1046 -0.0775 0.0000 0.0138
PMON L AmygCON PDS Yth 6644 13 Bpath 0.0029 -0.0550 -0.0261 0.0780 0.0148
PMON L AmygCON PDS Yth 6644 13 Direct 0.0636 0.0098 0.0367 0.0075 0.0137
PMON L AmygCON PDS Yth 6644 13 Indirect 0.0044 -0.0003 0.0020 0.0945 0.0012
PMON L AmygCON PDS Yth 6644 13 Total 0.0653 0.0122 0.0388 0.0043 0.0136
PMON R AmygCON PDS ParYth 6644 13 Apath -0.0528 -0.1031 -0.0779 0.0000 0.0128
PMON R AmygCON PDS ParYth 6644 13 Bpath -0.0281 -0.0901 -0.0591 0.0002 0.0158
PMON R AmygCON PDS ParYth 6644 13 Direct 0.0408 -0.0125 0.0142 0.2978 0.0136
PMON R AmygCON PDS ParYth 6644 13 Indirect 0.0075 0.0017 0.0046 0.0019 0.0015
PMON R AmygCON PDS ParYth 6644 13 Total 0.0452 -0.0077 0.0188 0.1648 0.0135
PMON R AmygCON PDS Par 6644 13 Apath -0.0251 -0.0687 -0.0469 0.0000 0.0111
PMON R AmygCON PDS Par 6644 13 Bpath -0.0501 -0.1072 -0.0786 0.0000 0.0146
PMON R AmygCON PDS Par 6644 13 Direct 0.0416 -0.0107 0.0155 0.2459 0.0133
PMON R AmygCON PDS Par 6644 13 Indirect 0.0059 0.0015 0.0037 0.0011 0.0011
PMON R AmygCON PDS Par 6644 13 Total 0.0453 -0.0070 0.0192 0.1507 0.0133
PMON R AmygCON PDS Yth 6644 13 Apath -0.0512 -0.1042 -0.0777 0.0000 0.0135
PMON R AmygCON PDS Yth 6644 13 Bpath 0.0043 -0.0532 -0.0245 0.0953 0.0147
PMON R AmygCON PDS Yth 6644 13 Direct 0.0427 -0.0092 0.0167 0.2059 0.0132
PMON R AmygCON PDS Yth 6644 13 Indirect 0.0042 -0.0004 0.0019 0.1100 0.0012
PMON R AmygCON PDS Yth 6644 13 Total 0.0445 -0.0072 0.0187 0.1576 0.0132

As we see, the table length for a single path type, such as Direct effect, is 135.

5.5 Specr Plots

For these plots, as is used in prior work, we report the ordered effects across out multiverse of indirect, direct and total effects. There are some additional inferential procedures we can consider, see Dani Cosme’s Documentation example. However, these bootstrapped estimates are typically done for simple X~Y models and would require tailoring for the mediation. Some other representations for specr models:

Of note, while there are no significant effects here (conf.high/conf/low overlap with 0), these plots will color points blue/red for p < .05 for post/neg effects.

How to interpret the specification curve plots? The multi-panel figures represent the effect size and significance in Panel A and the variables (significance denoted blue/red and null denoted gray) used in the models in Panel B. The interpretation of these panels are straightforward, the significant negative effect (red) in panel B, will have an associated X (predictor), M (mediator) and Y (outcome) denoted in red that was fit for that model. All of these models also controlled for age, sex, and race.

Like in Rinjhart 2021, we create a specification curve figures for the (1) Direct effect, (2) Indirect effect and (3) Total effect. For each figure, the effects are ordered, from least to greatist, thus, the order of variables/effects across these three figures will not always overlap.

5.5.1 Direct Effect

specr_direct <- mediation_multiverse %>% 
  filter(Effect == "Direct")
specr_direct$controls <- "Age + Race + Sex"


plot_a <- plot_curve(df = specr_direct, 
                     ci = TRUE, desc = FALSE, legend = FALSE, null = 0) +
  labs(caption = "Blue = Significant Positive (p < .05); Red = Significant Negative (p < .05)")

plot_b <- plot_choices(df = specr_direct, choices = c("X", "Y", "M"), desc = F, null = 0) +
  labs(y = "Variables", x = "Ordered Specification Curve \n Direct Effect")

#jpeg("Figures/Aim2/DirectEffect_Multiverse.jpeg", units = "in",
#     width = 10, height = 5, res = 300)

cowplot::plot_grid(plot_a, plot_b, ncol = 1, align = "v", axis = 'tblr',
                       labels = c('A', 'B'), rel_heights = c(1, 2),
                   label_fontfamily = "Times", label_size = 12)

#dev.off()

5.5.2 Indirect Effect

specr_indirect <- mediation_multiverse %>% 
  filter(Effect == "Indirect")
specr_indirect$controls <- "Age + Race + Sex"

plot_a_ind <- plot_curve(df = specr_indirect, 
                     ci = TRUE, desc = F, legend = F) +
  labs(caption = "Blue = Significant Positive (p < .05); Red = Significant Negative (p < .05)")

plot_b_ind <- plot_choices(df = specr_indirect, choices = c("X", "Y", "M"), desc = F) +
  labs(y = "Variables", x = "Ordered Specification Curve \n Indirect Effect")


#jpeg("Figures/Aim2/IndirectEffect_Multiverse.jpeg", units = "in",
#    width = 10, height = 5, res = 300)

cowplot::plot_grid(plot_a_ind, plot_b_ind, ncol = 1, align = "v", axis = 'tblr',
                       labels = c('A', 'B'), rel_heights = c(1, 2),
                   label_fontfamily = "Times", label_size = 12)

#dev.off()

5.5.3 Total Effect

specr_total <- mediation_multiverse %>% 
  filter(Effect == "Total")

specr_total$controls <- "Age + Race + Sex"

plot_a_tot <- plot_curve(df = specr_total, 
                     ci = TRUE, desc = F, legend = T) +
  labs(caption = "Blue = Significant Positive (p < .05); Red = Significant Negative (p < .05)")


plot_b_tot <- plot_choices(df = specr_total, choices = c("X", "Y", "M"), desc = F) +
  labs(y = "Variables", x = "Ordered Specification Curve \n Total Effect")

#jpeg("Figures/Aim2/TotalEffect_Multiverse.jpeg", units = "in",
#     width = 10, height = 5, res = 300)

cowplot::plot_grid(plot_a_tot, plot_b_tot, ncol = 1, align = "v", axis = 'tblr',
                       labels = c('A', 'B'), rel_heights = c(1, 2),
                   label_fontfamily = "Times", label_size = 12)

#dev.off()

5.5.4 A/B path

specr_apth <- mediation_multiverse %>% 
  filter(Effect == "Apath")
specr_apth$controls <- "Age + Race + Sex"


plot_a <- plot_curve(df = specr_apth, 
                     ci = TRUE, desc = F, legend = F) +
  labs(caption = "Blue = Significant Positive (p < .05); Red = Significant Negative (p < .05)")

plot_b <- plot_choices(df = specr_apth, choices = c("X", "Y", "M"), desc = F) +
  labs(y = "Variables", x = "Ordered Specification Curve \n A-path Effect")

#jpeg("Figures/Aim2/Apath_Multiverse.jpeg", units = "in",
#     width = 10, height = 5, res = 300)

cowplot::plot_grid(plot_a, plot_b, ncol = 1, align = "v", axis = 'tblr',
                       labels = c('A', 'B'), rel_heights = c(1, 2),
                   label_fontfamily = "Times", label_size = 12)

#dev.off()
specr_bpth <- mediation_multiverse %>% 
  filter(Effect == "Bpath")
specr_bpth$controls <- "Age + Race + Sex"


plot_a <- plot_curve(df = specr_bpth, 
                     ci = TRUE, desc = F, legend = F) +
  labs(caption = "Blue = Significant Positive (p < .05); Red = Significant Negative (p < .05)")

plot_b <- plot_choices(df = specr_bpth, choices = c("X", "Y", "M"), desc = F) +
  labs(y = "Variables", x = "Ordered Specification Curve \n B-path Effect")

#jpeg("Figures/Aim2/Bpath_Multiverse.jpeg", units = "in",
#    width = 10, height = 5, res = 300)

cowplot::plot_grid(plot_a, plot_b, ncol = 1, align = "v", axis = 'tblr',
                       labels = c('A', 'B'), rel_heights = c(1, 2),
                   label_fontfamily = "Times", label_size = 12)

#dev.off()

5.5.5 Within Path by PDS type Sig %

mediation_multiverse %>% 
  dplyr::group_by(Effect, M) %>% 
  mutate(N_Effect = n()) %>% 
  filter(p.value <= .05)  %>% 
  dplyr::summarize(count_sig = n()) %>% 
  group_by(Effect, M) %>% 
  dplyr::summarize(perc_sig = round((count_sig/45),2), count_sig)
## # A tibble: 15 × 4
## # Groups:   Effect [5]
##    Effect   M          perc_sig count_sig
##    <chr>    <chr>         <dbl>     <int>
##  1 Apath    PDS Par        1           45
##  2 Apath    PDS ParYth     1           45
##  3 Apath    PDS Yth        0.78        35
##  4 Bpath    PDS Par        0.78        35
##  5 Bpath    PDS ParYth     0.38        17
##  6 Bpath    PDS Yth        0.04         2
##  7 Direct   PDS Par        0.42        19
##  8 Direct   PDS ParYth     0.42        19
##  9 Direct   PDS Yth        0.44        20
## 10 Indirect PDS Par        0.78        35
## 11 Indirect PDS ParYth     0.38        17
## 12 Indirect PDS Yth        0.04         2
## 13 Total    PDS Par        0.44        20
## 14 Total    PDS ParYth     0.44        20
## 15 Total    PDS Yth        0.44        20

5.6 Plotting Multiverse within factors/measures

5.6.1 Within Factor Estimates

Here we provide an example of how estimates can vary across factor derived scores across the 5 DVs for the parent reported PDS scores. Below, we plot the beta estimate + 95% CI for family factor (filled square) for each of the 5 DVs, In addition to this, we include the beta estimates for each o the 5 DVs for the parent reported PDS and the following factors:

  • Parent Factor (Circle)
  • Demographic Factor (Triangle)
  • Child Factor (X)

These helps visualize how much a variable can vary in it’s indirect & direct effect w/ between the IV (environmental factors) and DVs (brain outcomes)

color_1 <- cbPalette <- c("black", "chocolate1", "firebrick", "slateblue", "turquoise4")# six colours for six factors

#jpeg("Figures/Aim2/WithinFactor_betas.jpeg", units = "in",
#     width = 10, height = 6, res = 250)

mediation_multiverse %>% 
  ggplot(aes(colour = Y)) +
  geom_errorbar(aes(x = Y, y = estimate,
                    ymin = conf.low, ymax = conf.high), 
                 width = .1, alpha = .5, 
                data = . %>%  filter(X %in% "FamEnv Fact", M %in% "PDS Par")) +
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 15, # filled square
             data = . %>%  filter(X %in% "FamEnv Fact", M %in% "PDS Par")) + # subset data for factor + par puberty
  
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 16, # circle
             position = position_jitterdodge(),
             data = . %>%  filter(X %in% "Par Fact", M %in% "PDS Par")) + # subset data for factor + par puberty
  
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 17, # triangle
             position = position_jitterdodge(),
             data = . %>%  filter(X %in% "Demo Fact", M %in% "PDS Par")) + # subset data for factor + par puberty
  
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 18, # X
             position = position_jitterdodge(),
             data = . %>%  filter(X %in% "Child Fact", M %in% "PDS Par")) + # subset data for factor + par puberty
  
  labs(y = "Beta Estimates", 
       colour = "Brain Outcome") +
  scale_colour_manual(values = color_1)+
  facet_wrap(~Effect, scale = "free")+
  theme_minimal()+
  theme(axis.title.x = element_blank(), axis.ticks.x = element_blank(), 
        axis.line.x = element_blank(), axis.text.x = element_blank())

#dev.off()

5.6.2 Within Measure Estimates

Here we provide an example of how estimates can vary across individual measures across the 5 DVs for the parent reported PDS scores. Below, we plot the beta estimate + 95% CI for family factor (filled square) for each of the 5 DVs, In addition to this, we include the beta estimates for each of the 5 DVs for the parent reported PDS and the following measures:

  • FES Youth
  • FES Parent
  • Parental Monitoring
  • Parental Acceptance
  • Avg Income/Education

“FES_youth”,“FES_parent”, “PMON”,“CRPBI”,“Avg_IncomeEduc”

These helps visualize how much a variable can vary in it’s indirect & direct effect w/ between the IV (environmental factors) and DVs (brain outcomes)

color_1 <- cbPalette <- c("black", "chocolate1", "firebrick", "slateblue", "turquoise4")# six colours for six factors

#jpeg("Figures/Aim2/WithinMeasure_betas.jpeg", units = "in",
#     width = 10, height = 6, res = 250)

mediation_multiverse %>% 
  ggplot(aes(colour = Y)) +
  geom_errorbar(aes(x = Y, y = estimate,
                    ymin = conf.low, ymax = conf.high), 
                 width = .1, alpha = .5, 
                data = . %>%  filter(X %in% "FamEnv Fact", M %in% "PDS Par")) +
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 15, # filled square
             data = . %>%  filter(X %in% "FamEnv Fact", M %in% "PDS Par")) + # subset data for factor + par puberty
  
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 1, # circle
             position = position_jitterdodge(),
             data = . %>%  filter(X %in% "FES Yth", M %in% "PDS Par")) + # subset data for FES youth + par puberty
  
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 3, # +
             position = position_jitterdodge(),
             data = . %>%  filter(X %in% "FES Par", M %in% "PDS Par")) + # subset data for FES parent + par puberty
  
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 4, # x
             position = position_jitterdodge(),
             data = . %>%  filter(X %in% "PMON", M %in% "PDS Par")) + # subset data for Parental monitoring + par puberty
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 5, # diamond
             position = position_jitterdodge(),
             data = . %>%  filter(X %in% "Par Accept", M %in% "PDS Par")) + # subset data for Parental acceptance + par puberty
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 8, # star
             position = position_jitterdodge(),
             data = . %>%  filter(X %in% "Avg IncEdu", M %in% "PDS Par")) + # subset data for Avg Inc/Edu + par puberty
  
  labs(y = "Beta Estimates", 
       colour = "Brain Outcome") +
  scale_colour_manual(values = color_1)+
  facet_wrap(~Effect, scale = "free")+
  theme_minimal()+
  theme(axis.title.x = element_blank(), axis.ticks.x = element_blank(), 
        axis.line.x = element_blank(), axis.text.x = element_blank())

#dev.off()

Reduced to only indirect & direct effects

IndDir_A <- mediation_multiverse %>% 
  filter(Effect == "Direct" | Effect == "Indirect") %>% 
  ggplot(aes(colour = Y)) +
  geom_errorbar(aes(x = Y, y = estimate,
                    ymin = conf.low, ymax = conf.high), 
                 width = .1, alpha = .5, 
                data = . %>%  filter(X %in% "FamEnv Fact", M %in% "PDS Par", 
                                  Effect %in% c("Direct", "Indirect"))) +
  
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 15, # filled square
             data = . %>%  filter(X %in% "FamEnv Fact", M %in% "PDS Par", 
                                  Effect %in% c("Direct", "Indirect"))) + # subset data for factor + par puberty
  
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 16, # circle
             position = position_jitterdodge(), 
             data = . %>%  filter(X %in% "Par Fact", M %in% "PDS Par", 
                                  Effect %in% c("Direct", "Indirect"))) + # subset data for factor + par puberty
  
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 17, # triangle
             position = position_jitterdodge(), 
             data = . %>%  filter(X %in% "Demo Fact", M %in% "PDS Par", 
                                  Effect %in% c("Direct", "Indirect"))) + # subset data for factor + par puberty
  
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 18, # diamond
             position = position_jitterdodge(), 
             data = . %>%  filter(X %in% "Child Fact", M %in% "PDS Par", 
                                  Effect %in% c("Direct", "Indirect"))) + # subset data for factor + par puberty
  
  labs(y = "Beta Estimates", 
       colour = "Brain Outcome"
       #caption = "Square = FamEnv Factor; Circle: Parent Factor \n Triangle = Demographic Factor; X = Child Factor"
       ) +
  scale_colour_manual(values = color_1)+
  facet_wrap(~Effect, scale = "free")+
  theme_minimal()+
  theme(axis.title.x = element_blank(), axis.ticks.x = element_blank(), 
        axis.line.x = element_blank(), axis.text.x = element_blank())



IndDir_B <- mediation_multiverse %>% 
  filter(Effect == "Direct" | Effect == "Indirect") %>% 
  ggplot(aes(colour = Y)) +
  geom_errorbar(aes(x = Y, y = estimate,
                    ymin = conf.low, ymax = conf.high), 
                 width = .1, alpha = .5, 
                data = . %>%  filter(X %in% "FamEnv Fact", M %in% "PDS Par")) +
  
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 15, # filled square
             data = . %>%  filter(X %in% "FamEnv Fact", M %in% "PDS Par", 
                                  Effect %in% c("Direct", "Indirect"))) + # subset data for factor + par puberty
  
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 1, # circle
             position = position_jitterdodge(),
             data = . %>%  filter(X %in% "FES Yth", M %in% "PDS Par", 
                                  Effect %in% c("Direct", "Indirect"))) + # subset data for FES youth + par puberty
  
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 3, # +
             position = position_jitterdodge(),
             data = . %>%  filter(X %in% "FES Par", M %in% "PDS Par", 
                                  Effect %in% c("Direct", "Indirect"))) + # subset data for FES parent + par puberty
  
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 4, # x
             position = position_jitterdodge(),
             data = . %>%  filter(X %in% "PMON", M %in% "PDS Par", 
                                  Effect %in% c("Direct", "Indirect"))) + # subset data for Parental monitoring + par puberty
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 5, # diamond
             position = position_jitterdodge(),
             data = . %>%  filter(X %in% "Par Accept", M %in% "PDS Par", 
                                  Effect %in% c("Direct", "Indirect"))) + # subset data for Parental acceptance + par puberty
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 8, # star
             position = position_jitterdodge(),
             data = . %>%  filter(X %in% "Avg IncEdu", M %in% "PDS Par", 
                                  Effect %in% c("Direct", "Indirect"))) + # subset data for Avg Inc/Edu + par puberty
  
  labs(y = "Beta Estimates", 
       colour = "Brain Outcome"
       #caption = "Square = FamEnv Factor; Circle = FES Yth;  Plus = FES; \n X = PMON; Diamond = CRPBI; Asterisk = Avg Inc/Edu"
       ) +
  scale_colour_manual(values = color_1)+
  facet_wrap(~Effect, scale = "free")+
  theme_minimal()+
  theme(axis.title.x = element_blank(), axis.ticks.x = element_blank(), 
        axis.line.x = element_blank(), axis.text.x = element_blank(),
        legend.position = "none")


#jpeg("Figures/Aim2/WithinFactMeasure_DirInd.jpeg", units = "in",
#     width = 9, height = 5, res = 300)

#IndDir_A / IndDir_B + plot_layout(guides = 'collect')

cowplot::plot_grid(IndDir_A, IndDir_B, ncol = 1, align = "v", axis = 'tblr',
                       labels = c('A', 'B'), rel_heights = c(1, 1),
                   label_fontfamily = "Times", label_size = 14)

#dev.off()

Reduced to rsfmri to visual differences in closely clustered data that are scaled diff in structural estimates

color_2 <- cbPalette <- c("slateblue", "turquoise4")

IndDir_A_rsf <- mediation_multiverse %>% 
  filter(Y == "R AmygCON" | Y == "L AmygCON") %>% 
  filter(Effect == "Direct" | Effect == "Indirect") %>% 
  ggplot(aes(colour = Y)) +
  geom_errorbar(aes(x = Y, y = estimate,
                    ymin = conf.low, ymax = conf.high), 
                 width = .1, alpha = .5, 
                data = . %>%  filter(X %in% "FamEnv Fact", M %in% "PDS Par", 
                                  Effect %in% c("Direct", "Indirect"))) +
  
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 15, # filled square
             data = . %>%  filter(X %in% "FamEnv Fact", M %in% "PDS Par", 
                                  Effect %in% c("Direct", "Indirect"))) + # subset data for factor + par puberty
  
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 16, # circle
             position = position_jitterdodge(), 
             data = . %>%  filter(X %in% "Par Fact", M %in% "PDS Par", 
                                  Effect %in% c("Direct", "Indirect"))) + # subset data for factor + par puberty
  
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 17, # triangle
             position = position_jitterdodge(), 
             data = . %>%  filter(X %in% "Demo Fact", M %in% "PDS Par", 
                                  Effect %in% c("Direct", "Indirect"))) + # subset data for factor + par puberty
  
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 18, # diamond
             position = position_jitterdodge(), 
             data = . %>%  filter(X %in% "Child Fact", M %in% "PDS Par", 
                                  Effect %in% c("Direct", "Indirect"))) + # subset data for factor + par puberty
  
  labs(y = "Beta Estimates", 
       colour = "Brain Outcome"
       #caption = "Square = FamEnv Factor; Circle: Parent Factor \n Triangle = Demographic Factor; X = Child Factor"
       ) +
  scale_colour_manual(values = color_2)+
  facet_wrap(~Effect, scale = "free")+
  theme_minimal()+
  theme(axis.title.x = element_blank(), axis.ticks.x = element_blank(), 
        axis.line.x = element_blank(), axis.text.x = element_blank())



IndDir_B_rsf <- mediation_multiverse %>% 
  filter(Y == "R AmygCON" | Y == "L AmygCON") %>% 
  filter(Effect == "Direct" | Effect == "Indirect") %>% 
  ggplot(aes(colour = Y)) +
  geom_errorbar(aes(x = Y, y = estimate,
                    ymin = conf.low, ymax = conf.high), 
                 width = .1, alpha = .5, 
                data = . %>%  filter(X %in% "FamEnv Fact", M %in% "PDS Par")) +
  
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 15, # filled square
             data = . %>%  filter(X %in% "FamEnv Fact", M %in% "PDS Par", 
                                  Effect %in% c("Direct", "Indirect"))) + # subset data for factor + par puberty
  
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 1, # circle
             position = position_jitterdodge(),
             data = . %>%  filter(X %in% "FES Yth", M %in% "PDS Par", 
                                  Effect %in% c("Direct", "Indirect"))) + # subset data for FES youth + par puberty
  
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 3, # +
             position = position_jitterdodge(),
             data = . %>%  filter(X %in% "FES Par", M %in% "PDS Par", 
                                  Effect %in% c("Direct", "Indirect"))) + # subset data for FES parent + par puberty
  
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 4, # x
             position = position_jitterdodge(),
             data = . %>%  filter(X %in% "PMON", M %in% "PDS Par", 
                                  Effect %in% c("Direct", "Indirect"))) + # subset data for Parental monitoring + par puberty
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 5, # diamond
             position = position_jitterdodge(),
             data = . %>%  filter(X %in% "Par Accept", M %in% "PDS Par", 
                                  Effect %in% c("Direct", "Indirect"))) + # subset data for Parental acceptance + par puberty
  geom_point(aes(x = Y, y = estimate), fill = "white", shape = 8, # star
             position = position_jitterdodge(),
             data = . %>%  filter(X %in% "Avg IncEdu", M %in% "PDS Par", 
                                  Effect %in% c("Direct", "Indirect"))) + # subset data for Avg Inc/Edu + par puberty
  
  labs(y = "Beta Estimates", 
       colour = "Brain Outcome"
       #caption = "Square = FamEnv Factor; Circle = FES Yth;  Plus = FES; \n X = PMON; Diamond = CRPBI; Asterisk = Avg Inc/Edu"
       ) +
  scale_colour_manual(values = color_2)+
  facet_wrap(~Effect, scale = "free")+
  theme_minimal()+
  theme(axis.title.x = element_blank(), axis.ticks.x = element_blank(), 
        axis.line.x = element_blank(), axis.text.x = element_blank(),
        legend.position = "none")


#jpeg("Figures/Aim2/WithinFactMeasure_rfMRIonly_DirInd.jpeg", units = "in",
#     width = 9, height = 5, res = 300)

cowplot::plot_grid(IndDir_A_rsf, IndDir_B_rsf, ncol = 1, align = "v", axis = 'tblr',
                       labels = c('A', 'B'), rel_heights = c(1, 1),
                   label_fontfamily = "Times", label_size = 14)

#dev.off()

5.7 Association different factors/items

To help with the interpretability/overlap of the effects in the above, we provide the pearson product correlations below. Some values may be opposite direction of others, such as Demographics and Family environment factors.

We summarize these correlations in a table and a shaded plot below. Some information about how to use the plotting function, see the corrplot information. The plot includes significance values for each correlation:

  • < 05 ~ *
  • < .01 ~ **
  • < .001 ~ ***
  • correlations with sig values are > .05 are blank
var_list = data.frame(
  # Covariates
  "Age" = data$Age, "Sex" = data$sex_r,
  # IVS
  "FamEnv_Fact" = data$FamEnv_Fact, 
  "Demo_Fact" = data$Demo_Fact, 
  "Child_Fact" = data$Child_Fact, 
  "Par_Fact" = data$Par_Fact,
  "FES_Yth" = data$FES_youth,"FES_Par" = data$FES_parent, 
  "PMON" = data$PMON,"Par_Accept" = data$CRPBI, "Avg_IncEdu" = data$Avg_IncomeEduc,
  # Mediators
  "Par_PDS" = data$p_puberty,"Yth_PDS" = data$y_puberty,"ParYth_PDS" = data$avg_puberty,
  # Dependent variables
  "Amyg_Vol" = data$Amygdala_vol,
  "ACC_CT" = data$ACC_CT,"ACC_CA" = data$ACC_CA,
  "L_AmygCON" = data$L_AmygCing_rest, "R_AmygCON" = data$R_AmygCing_rest
)
  


#Var_corr = cor(var_list, use = "pairwise", method = "pearson")
Var_corr = rcorr(as.matrix(var_list), type = "pearson")

# rename column/row names for r-values to updated printed table

colnames(Var_corr$r) <- c("1","2","3","4","5","6","7","8","9","10","11","12",
                          "13","14","15","16","17","18","19")
rownames(Var_corr$r) <- c("1. Age", "2. Sex", "3. FamEnv Fact", 
                         "4. Demo Fact", "5. Child Fact", "6, Par Fact",
                         "7. FES Yth", "8. FES Par", "9. PMON", "10. Par Accept",
                         "11. Avg IncEdu", "12. PDS Par", "13. PDS Yth",
                         "14. PDS ParYth", "15. Amyg Vol", "17. ACC CT",
                         "17. ACC CA", "18. L AmygCON", "19. R AmygCON")

knitr::kable(Var_corr$r, digits = 2) %>% 
  kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive"),
                full_width = F, html_font = "Times")
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
  1. Age
1.00 0.03 0.04 0.01 0.07 -0.01 0.06 -0.01 0.10 0.01 0.01 0.18 0.12 0.18 0.08 -0.17 0.14 0.00 0.00
  1. Sex
0.03 1.00 -0.08 0.01 -0.15 -0.04 -0.06 -0.03 -0.17 -0.05 0.01 -0.48 -0.19 -0.42 0.14 -0.06 0.04 0.03 -0.01
  1. FamEnv Fact
0.04 -0.08 1.00 0.64 0.59 0.70 0.54 0.64 0.41 0.40 0.59 -0.13 -0.13 -0.14 -0.02 -0.01 -0.01 0.11 0.09
  1. Demo Fact
0.01 0.01 0.64 1.00 0.13 0.13 0.15 0.09 0.15 0.06 0.95 -0.24 -0.17 -0.23 0.00 0.01 -0.04 0.16 0.12
  1. Child Fact
0.07 -0.15 0.59 0.13 1.00 0.13 0.57 0.12 0.70 0.81 0.12 0.04 -0.07 -0.01 -0.01 -0.03 0.04 0.04 0.04
6, Par Fact -0.01 -0.04 0.70 0.13 0.13 1.00 0.22 0.94 0.04 0.08 0.11 -0.04 -0.02 -0.03 -0.02 0.00 -0.01 0.03 0.01
  1. FES Yth
0.06 -0.06 0.54 0.15 0.57 0.22 1.00 0.20 0.24 0.29 0.14 0.00 -0.08 -0.04 0.00 -0.01 0.00 0.04 0.05
  1. FES Par
-0.01 -0.03 0.64 0.09 0.12 0.94 0.20 1.00 0.05 0.07 0.07 -0.03 0.00 -0.02 -0.02 0.00 0.00 0.02 0.01
  1. PMON
0.10 -0.17 0.41 0.15 0.70 0.04 0.24 0.05 1.00 0.36 0.14 0.06 -0.03 0.03 -0.02 -0.04 0.03 0.03 0.02
  1. Par Accept
0.01 -0.05 0.40 0.06 0.81 0.08 0.29 0.07 0.36 1.00 0.05 -0.02 -0.05 -0.04 0.00 -0.01 0.04 0.03 0.03
  1. Avg IncEdu
0.01 0.01 0.59 0.95 0.12 0.11 0.14 0.07 0.14 0.05 1.00 -0.19 -0.15 -0.19 0.00 0.03 -0.04 0.14 0.12
  1. PDS Par
0.18 -0.48 -0.13 -0.24 0.04 -0.04 0.00 -0.03 0.06 -0.02 -0.19 1.00 0.43 0.85 -0.04 -0.04 0.04 -0.09 -0.06
  1. PDS Yth
0.12 -0.19 -0.13 -0.17 -0.07 -0.02 -0.08 0.00 -0.03 -0.05 -0.15 0.43 1.00 0.84 -0.01 -0.01 0.01 -0.04 -0.03
  1. PDS ParYth
0.18 -0.42 -0.14 -0.23 -0.01 -0.03 -0.04 -0.02 0.03 -0.04 -0.19 0.85 0.84 1.00 -0.03 -0.02 0.02 -0.08 -0.06
  1. Amyg Vol
0.08 0.14 -0.02 0.00 -0.01 -0.02 0.00 -0.02 -0.02 0.00 0.00 -0.04 -0.01 -0.03 1.00 -0.04 0.08 0.01 -0.02
  1. ACC CT
-0.17 -0.06 -0.01 0.01 -0.03 0.00 -0.01 0.00 -0.04 -0.01 0.03 -0.04 -0.01 -0.02 -0.04 1.00 -0.44 -0.01 0.02
  1. ACC CA
0.14 0.04 -0.01 -0.04 0.04 -0.01 0.00 0.00 0.03 0.04 -0.04 0.04 0.01 0.02 0.08 -0.44 1.00 -0.02 -0.03
  1. L AmygCON
0.00 0.03 0.11 0.16 0.04 0.03 0.04 0.02 0.03 0.03 0.14 -0.09 -0.04 -0.08 0.01 -0.01 -0.02 1.00 0.57
  1. R AmygCON
0.00 -0.01 0.09 0.12 0.04 0.01 0.05 0.01 0.02 0.03 0.12 -0.06 -0.03 -0.06 -0.02 0.02 -0.03 0.57 1.00 
#jpeg("Figures/Descriptives/Correlations_Vars.jpeg", units = "in",
#     width = 10, height = 6, res = 250)
corrplot(Var_corr$r, type = "upper", 
         method =  "color", #title = "Pairwise Pearson r among variables", 
         tl.cex = 0.7, tl.col = 'black',#order = 'AOE', 
         p.mat = Var_corr$P,
         sig.level = c(0.001, 0.01, 0.05), pch.cex = 0.6,
         insig = 'label_sig', pch.col = 'black',
         mar=c(1, 1, 1, 1))

#dev.off()

5.7.1 Subsample Models to Validate Results

Selecting a random subsample of 135 models to validate

set.seed(1989)

subsample_models <- mediation_multiverse %>% 
  filter(Effect == "Indirect") %>%  # we are looking to subsample only our 135 unique models, not all effect types
  dplyr::select(X, Y, M) %>% 
  sample_n(tbl = .,
           size = 6,
           replace = FALSE)

write.csv(x = subsample_models, 
          file = "./subsample_N6_models.csv", 
          row.names = F)

Saving dataset to validate models on, e.g., simply selecting primary variables from the 128 variable list. This data will be used for coauthor, FH, to validate the random subsample of 3

sub_data <- data %>% 
  # Covariates
  dplyr::select(subjectkey, Age, sex_r, race_r,
  # IVS
  FamEnv_Fact, Demo_Fact, Child_Fact, Par_Fact, FES_youth, FES_parent, 
  PMON, CRPBI, Avg_IncomeEduc,
  # Mediators
  p_puberty, y_puberty,avg_puberty,
  # Dependent variables
  Amygdala_vol, ACC_CT, ACC_CA, L_AmygCing_rest, R_AmygCing_rest)

write.csv(x = sub_data, 
          file = "./sub_data.csv", 
          row.names = F)